J Neurosci Res

Dynamic Changes in the ExpressionPattern of Ecto-50-Nucleotidase in the RatModel of Cortical Stab Injury

Ivana Bjelobaba,1 Ana Parabucki,1 Irena Lavrnja,1 Danijela Stojkov,1 Sanja Dacic,2

Sanja Pekovic,1 Ljubisa Rakic,3 Mirjana Stojiljkovic,1,2 and Nadezda Nedeljkovic2*1Department for Neurobiology, Institute for Biological Research Sinisa Stankovic, University Belgrade,Belgrade, Serbia2Institute for Physiology and Biochemistry, Faculty of Biology, University Belgrade, Belgrade, Serbia3Serbian Academy of Sciences and Arts, Belgrade, Serbia

Traumatic injury induces massive release of ATP in theextracellular space, where it influences numerousaspects of neuronal, astrocytic, and microglialresponses to injury by activating P2X and P2Y recep-tors. The extracellular ATP actions are controlled by theectonucleotidase enzyme pathway, which hydrolysesATP to adenosine at all neuronal and nonneuronal celltypes. Adenosine activates its P1 receptors, whichhave important neuroprotective roles. The rate-limitingenzyme in the ectonucleotidase pathway is ecto-50-nu-cleotidase (e-5NT), which catalyzes the final step of de-phosphorylation of AMP to adenosine. The aim of thepresent study was to characterize the expressionpattern and cellular distribution of e-5NT in the perile-sioned cortex at 4 hr and 1, 2, 7, and 15 days after uni-lateral cortical stab injury (CSI). Immunoblot andimmunohistochemical studies showed that overalle-5NT expression was lower 4 hr and 1 day postinjuryand then gradually increased above the control levels.Double-immunofluorescence studies further showed incontrol tissue the presence of the enzyme in the mem-branes surrounding neuronal somata and apical den-drites and less frequently in astrocytes. CSI caused arapid (after 4 hr) and irreversible loss of the enzymefrom neurons, accounting for a decrease in the overallenzyme expression. This was accompanied with agradual increase in e-5NT-positive astrocytes, account-ing for up-regulation of the enzyme levels in the injuredarea. Thus, CSI induced dynamic changes in theexpression pattern of e-5NT that modify the ATP/aden-osine ratio and the extent of P1 and P2 receptors acti-vation and, therefore, outcome of the pathologicalprocesses after CSI. VVC 2011 Wiley-Liss, Inc.

Key words: brain; ecto-50-nucleotidase; purinergicsystems; ATP; astrocytes

Traumatic brain injury consists of primary injurythat is the consequence of immediate mechanical damageand secondary injury that evolves over a period ofminutes to days, causes pathological changes over a

much greater area, and is often responsible for many ofthe long-lasting neurological deficits (Chen et al., 2003).Brain injury induces release of ATP and other purinenucleotides (Melani et al., 2005; Franke et al., 2006b),which are important players in both secondary damageand repair, because they induce or contribute to manyaspects of astrocytic, microglial, and neuronal responsesto injury. This includes rapid transition from resting toactivated microglial cells (Davalos et al., 2005; Haynes etal., 2006; Xiang et al., 2006; Ohsawa et al., 2007), theirfunctional changes (Koizumi et al., 2007), and release ofcytokines (Di Virgilio et al., 2009; Takenouchi et al.,2009).

ATP exerts its effects via ionotropic (P2X) andmetabotropic (P2Y) receptors. By activating P2X recep-tors, ATP increases intracellular calcium (Inoue et al.,1995; Edwards, 1996) and induces neuronal cell loss(Amadio et al., 2002; Franke et al., 2006a). Via P2Yreceptors, ATP induces proliferation, stellation, andmigration of astrocytes (Wang et al., 2005; Neary et al.,2006) as well as release of astrocytic growth-promotingand -inhibiting factors (Ridet et al., 1997; Sofroniew,2005), resulting in the formation of a barrier betweenthe healthy and injured tissue. Extracellular ATP is rap-idly hydrolyzed either directly to AMP or to ADP,which is an agonist for some P2Y receptors. AMP ishydrolyzed adenosine, which acts at its own receptorfamily, coupled to either inhibition (A1 and A3) or acti-vation (A2A and A2B) of the adenylate cyclase signaling

Contract grant sponsor: Serbian Ministry of Science and Technology;

Contract grant number: 41014.

*Correspondence to: Nadezda Nedeljkovic, Department of General

Physiology and Biophysics, Institute for Physiology and Biochemistry,

Faculty of Biology, University Belgrade, Studentski trg 3, 11001

Belgrade, Serbia. E-mail: [email protected]

Received 25 October 2010; Revised 10 December 2010; Accepted 18

December 2010

Published online 17 February 2011 in Wiley Online Library

(wileyonlinelibrary.com). DOI: 10.1002/jnr.22599

Journal of Neuroscience Research 89:862–873 (2011)

' 2011 Wiley-Liss, Inc.

cascade (Cunha, 2005). In contrast to activation of P2Xreceptors, activation of the Gi/o signaling pathway by A1and A3 receptors leads to inhibition of neuronal firing,calcium influx, and neurotransmitter release, such as glu-tamate release, contributing to the neuroprotection(Stone et al., 2007). Activation of the Gs signaling path-way by A2A receptors enhances synaptic transmissionand glutamate release (O’Regan et al., 1992).

The availability of ATP, ADP, and adenosine andthe extent and duration of their actions at purinorecep-tors are controlled by the enzyme chain of ectonucleoti-dases, composed of several enzyme families differing intheir functional and molecular properties (Zimmermann,2000; Yegutkin, 2008). Ectonucleotidases include ecto-nucleoside triphospho-diphosphohydrolase (NTPDase)and ecto-nucleotide pyrophosphatase/phosphodiesterase(NPP) that hydrolyze ATP and ADP to AMP and ecto-50-nucleotidase (e-5NT; also known as CD73) that cata-lyzes the final step of dephosphorylation of ATP toadenosine. e-5NT is the key enzyme in this pathway,because its feed-forward inhibition by ATP and ADPcontrols the timing and rates of extracellular adenosineformation (James and Richardson, 1993). Thus, it is rea-sonable to suggest that changes in the expression andactivity of these enzymes are relevant for acute and sus-tained biochemical and cellular responses to injury.

Because of its role in the control of extracellularAMP and adenosine levels and consequently the activityof adenosine receptors, e-5NT plays important role inneuronal and glial cell functions. Previous studies havealso demonstrated significant up-regulation of e-5NT ac-tivity in damaged tissue in different models of brain injury(Nagy et al., 1997; Braun et al., 1998; Bonan et al.,2000a,b; Villa et al., 2002). With the rat model of corticalstab injury (SCI), we showed biphasic changes in e-5NTactivity, an immediate decrease in the activity (Nedelj-kovic et al., 2008), followed by a prominent up-regulationof the expression and activity of the enzyme that lasted for2 weeks after the injury (Nedeljkovic et al., 2006). Theaim of the present study was to characterize the expressionpattern of e-5NT in the perilesioned area of injured cortexat different postinjury times.

MATERIALS AND METHODS

Animals

Three-month-old male Wistar rats weighing between250 and 300 g at the time of surgery were used in this study.Animals were subjected to a 12-hr light/dark cycle, housedthree per cage, with free access to food and water. All animalswere treated in accordance with the principles from Guide forcare and use of laboratory animals (NIH Publication No. 80-23),and the protocols were approved by the Belgrade UniversityAnimal Care and Use Committee. All efforts were made tominimize the number of used animals and their suffering.

Surgical Procedure and CSI

Animals were anesthetized with diethyl ether and placedin the stereotaxic frame. The scalp was shaved, and an incision

was made along the midline of the scalp to expose bregma.CSI was inflicted as previously described (Nedeljkovic et al.,2006, 2008). Briefly, a 1-mm-wide dental drill was insertedvertically in the cranium on the left side at 2 mm caudal tobregma, 2 mm lateral to the midline, and 2 mm deep to thedura surface. The incision was closed with sutures. Anothergroup of animals was anesthetized and placed in the stereo-taxic frame and subjected to the same procedure, withoutcausing further damage to the skull (sham-operated animals).Animals of both groups were placed in a heated room andmonitored while recovering from anesthesia. Intact, age-matched animals were processed as controls. After recoveryfrom anesthesia, rats were returned to their home cages andallowed to survive for the indicated times (4 hr, 1 day, 2 days,7 days, 15 days). In total 66 animals (six per each experimentaland sham-operated group) were used in this study.

Tissue Preparation

Rats from each group (3/group) were decapitated at theindicated times after the stab injury or sham injury. Brains wereremoved, and ipsilateral cortices were dissected on ice andpooled. A crude membrane preparation was obtained essentiallyby following a procedure (Gray and Whittaker, 1962) previ-ously described (Nedeljkovic et al., 1998). The preparation wasisolated by tissue homogenization in the buffer containing320 mM sucrose and 10 mM Tris (pH 7.4), followed by cen-trifugation at 800g for 10 min at 48C. The resulting supernatantwas centrifuged at 9,000g for an additional 30 min, and the pel-leted membrane preparation was resuspended and homogenizedin 5 mM Tris, pH 7.4, prior to dilution in SDS-PAGE samplebuffer and immunoblot analysis. Protein content was deter-mined by the method of Markwell et al. (1978), and sampleswere kept on –708C until use.

Immunoblotting Procedure

Immunoblot analysis of e-5NT protein was performedas previously described (Nedeljkovic et al., 2005). Briefly,equal amounts of cortical tissue preparation (50 lg/lane) wereresolved by 7.5% SDS-PAGE gel according to Laemmli(1970) and transferred to a PVDF support membrane. Blotswere probed with goat polyclonal anti-e50-NT antibody(1:500 dilution; Santa Cruz Biotechnology, Santa Cruz, CA),and visualization was performed on X-ray films (Kodak) withthe use of chemiluminescence, after incubating the supportmembrane in donkey anti-goat IgG-horseradish peroxidase-conjugated secondary antibody (1:5,000 dilution; Santa CruzBiotechnology). In each blot, actin was used as a loading con-trol. Optical densities of e-5NT-immunoreactive bands werecalculated as arbitrary units in the Image Quant program afterlocal background subtraction. The results are expressed as theratio of e-5NT and actin optical densities relative to the valueobtained for intact control. Data presented in graphs are meanvalues 6 SEM, obtained from six immunoblots.

Immunohistochemistry

After overnight fixation in 4% paraformaldehyde in 0.1M phosphate buffer, pH 7.4, brains were cryoprotected ingraded sucrose (10–30% in 0.2 M phosphate buffer) at 48C.

Expression of e-5NT After Brain Injury 863

Journal of Neuroscience Research

Brains were frozen in 2-methyl butane and stored at –708Cuntil cryosectioning. Immunohistochemistry was performedon 16-lm-thick coronal sections. The antigen retrieval step inheated citrate buffer, pH 6, was performed to enhance stain-ing. Single immunoperoxidase labeling of e-5NT was per-formed according to the standard procedure. Nonspecificbinding was reduced with 5% donkey serum in 0.01 M phos-phate-buffered saline (PBS; pH7.4). Sections were incubatedovernight at 48C with polyclonal anti-e50-NT antibody (SantaCruz Biotechnology) at 1:100 dilution. Sections were thenincubated with biotinylated anti-goat IgG secondary antibody(1:200; Santa Cruz Biotechnology). Bound antibodies werevisualized with 3,30-diaminobenzidine (Dako, Carpinteria,CA) by the avidin-biotin peroxidase complex method, follow-ing standard protocols (Vector Laboratories, Burlingame, CA).After dehydratation and clearing, sections were mounted withDPX (Fluka). To test the specificity of the reaction, brain sec-tions were treated in the same way with the omission of theprimary antibody. Sections were examined and photographedwith ba Zeiss Axiovert microscope.

Double-immunofluorescence labeling was performed aspreviously described (Bjelobaba et al., 2010). After incubationin e-5NT antibody, secondary donkey anti-goat Alexa Fluor488 antibody (1:200 dilution; Invitrogen, Carlsbad, CA) wasapplied for 2 hr. After washing in PBS, sections were incubatedin the primary antibodies: mouse anti-NeuN or mouse antimi-crotubule-associated protein 2 (MAP-2; 1:100 dilution; Boeh-ringer). Secondary donkey anti-mouse Alexa Fluor 555 ordonkey anti-rabbit Alexa Fluor 555 (1:200 dilution; Invitrogen)was applied for 2 hr. All primary and secondary antibodies wereseparately applied. Some sections were counterstained withDAPI (Invitrogen). The sections were mounted in Mowiol(Calbiochem, La Jolla, CA) and examined under the ZeissAxiovert fluorescent microscope equipped with a camera. Theapotome system was used for obtaining optical sections.

RESULTS

To assess the pattern of e-5NT protein expressionfollowing stab injury, crude membrane fractions wereisolated from injured cortex at 4 hr, 1 day, 2 days, 7days, and 15 days postinjury. Immunoblot analysisshowed that e-5NT was present as a single band with amolecular mass of about 70 kDa (Fig. 1). There was asignificant postinjury decrease in protein level at 4 hr(27.6% 6 0.1%, P < 0.05) and 1 day (41.2% 6 0.1%, P

< 0.05) postinjury with respect to the intact control. At2 days postinjury, immunoreactivity had returned to thebaseline, followed by a gradual increase in the enzymelevel, which was significantly higher at 15 days postin-jury (37.7 6 0.1, P < 0.05) compared with the intactcontrol. These results are consistent with our previousstudy showing biphasic time-dependent changes in thetotal enzyme expression (Nedeljkovic et al., 2006,2008). Because only one immunoreactive band at themolecular weight expected for e-5NT was present, it isreasonable to suggest that the antibody used in this studycould also be useful for detecting the enzyme protein inbrain tissue.

To characterize further the injury-inducedchanges in e-5NT expression and to provide spatial in-formation, immunohistochemical study was performedon coronal sections obtained from control animals andanimals subjected to CSI at different postinjury times

Fig. 1. Quantitative immunoblot detection of e-5NT protein levelsin crude cortical membrane preparation isolated from the definedarea of injured cortex at different postinjury times. Bars representmean e-5NT protein abundance (6SEM) from six independentdeterminations expressed relative to the band intensity obtained forintact controls, arbitrarily defined as 1.0. Significance level is shownin the graph (*P < 0.05 vs. control), which is accompanied by a rep-resentative immunoblot.

Fig. 2. E-5NT immunoreactivity in rat cerebral cortex after corticalstab injury. Low-power micrographs of cerebral cortex in intact con-trol animals (A) and animals subjected to CSI at different postinjurytimes: 4 hr (D), 1 day (G), 2 days (I), 7 days (K) and 15 days (M)postinjury. High-power micrographs are magnified from the areaswith corresponding marks. In control sections (A–C), e-5NT immu-noreactivity is present in all cortical layers, mostly at neuronal cellbodies (B,C arrows) and in rare astrocytes (B, arrowhead). In sectionsobtained 4 hr postinjury (D), the general intensity of e-5NT signaldecreases, and only sporadic e-5NT1 neuronal cell bodies are evident(E,F, arrows). At 1 day postinjury (G), numerous e-5NT1 cell

bodies (H) with inconspicuous processes are seen. At 2 days postin-jury, immunosignal is present only in neuronal cell bodies in deepercortical layers (I) and in cells with radially oriented thin processesthat belong to astroglia (J). In sections obtained 7 days (K) and 14days (M) postinjury, e-5NT immunosignal increases, staining cellswith large cell bodies and thick radially oriented processes (L,N) thatbelong to reactive astroglia. In deeper cortical layers (O), a few scat-tered e-5NT1 neuronal cell bodies are also observed. Scale bars 5

100 lm in A (applies to A,D,G,I,K,M); 25 lm in B (applies toB,C,E,F,H,J,L,N,O). [Color figure can be viewed in the online issue,which is available at wileyonlinelibrary.com.]

864 Bjelobaba et al.


Figure 2.



(Fig. 2). In the control sections, enzyme immunoreac-tivity was present in all cortical layers (Fig. 2A), pre-dominantly at neuronal cell bodies (Fig. 2B,C) andcells with radially oriented thin processes that probablybelong to astroglia (Fig. 2B). Confirming what waspreviously depicted by immunoblotting, the general in-tensity of e-5NT immunoreactivity initially decreases

during the acute phase of injury, between 4 hr (Fig.2D–F) and 1 day (Fig. 2G,H) postinjury, and thengradually increases in sections obtained 2 days (Fig.2I,J), 7 days (Fig. 2K,L), and 15 days (Fig. 2M–O)postinjury. At 1 day (Fig. 2H) and 2 days (Fig. 2J)postinjury, immunosignal was present at neuronal cellbodies in deeper cortical layers. However, starting from

Fig. 3. Double-immunofluorescence staining with the antibodyagainst e-5NT (green fluorescence) and NeuN (red fluorescence). Incontrol sections (A–C), e-5NT immunofluorescence was detected atnumerous neuronal somata and apical dendrites. In sections obtained4 hr postinjury (D–F), the number of strongly labeled e-5NT1 neuro-nal cell bodies decreases, with a prominent lack of dendritic staining.At 1 day postinjury (G–I), only few scattered neuronal somata remain

e-5NT1. At 2 days postinjury (J–L), the general intensity of e-5NTimmunofluorescence increases, but the signal is only sporadically pres-ent at NeuN1 cell bodies. At 7 days (M–O) and 15 days (P–R) postin-jury, e-5NT immunosignal is localized at thick processes and largesomata that belong to reactive astrocytes. Scale bar 5 100 lm.[Color figure can be viewed in the online issue, which is available atwileyonlinelibrary.com.]



2 days postinjury and afterward (Fig. 2L,N), immuno-reactivity was increasingly present at large cells, pre-sumably reactive astroglia, based on morphological cri-

teria. In the lesion surroundings, e-5NT antibody spor-adically labeled beaded neuronal fibers, as previouslyreported (Bjelobaba et al., 2010).

Fig. 4. Fluorescence analysis for e-5NT (green fluorescence), MAP-2(red fluorescence) and DAPI (blue fluorescence). In the control sections(A–F), e-5NT signal is present at neuronal cell bodies and MAP-21 pri-mary dendrites, seen as fine, punctate staining at higher magnification

(D–F). At 1 day postinjury (G–L), aq drastic loss in both e-5NT andMAP-2 immunoreactivity is seen. Scale bars 5 10 lm In C (applies toA–C,G–I); 5 lm in F (applies to D–F,J–L). [Color figure can be viewedin the online issue, which is available at wileyonlinelibrary.com.]



Double-immunofluorescence staining with theantibody against e-5NT (Fig. 3, green fluorescence) andthe neuronal marker NeuN (Fig. 3, red fluorescence)was performed to characterize neuronal expression ofe-5NT. In intact controls (Fig. 3A–C), the strongest im-munoreactivity corresponding to e-5NT protein wasseen at the membranes surrounding neuronal somata andapical dendrites, whereas axonal processes leading fromthe cells were not visualized (Fig. 3B). Overall intensityof e-5NT immunofluorescence after the injury corre-sponded to changes revealed by immunohistochemistry.Starting from 4 hr postinjury, the most prominentchange in the expression pattern was an apparentdecrease in the number of e-5NT-positive neuronal cellbodies and an almost complete disappearance of e-5NTimmunofluorescence from apical dendrites (Fig. 3D–F).In 1-day-postinjury sections (Fig. 3G–I), e-5NT immu-nofluorescence had disappeared from neuronal somata,exhibiting a diffuse staining pattern (Fig. 3I). Analysis of

NeuN immunofluorescence also revealed a certain levelof neuronal cell loss in the perilesioned area as soon as 4hr postinjury (Fig. 3E). In 1-day-postinjury sections (Fig.3H), the extent of neuronal loss in the same area slightlyincreased and remained unchanged afterward. At 2 dayspostinjury (Fig. 2J–L), the general intensity of e-5NTimmunofluorescence increased (Fig. 3J). However,immunfluorescence was only sporadically present inNeuN-positive cell bodies (Fig. 3L). At 7 (Fig. 3M–O)and 14 (Fig. 3P–R) days postinjury, the e-5NT immu-nofluorescence was localized predominantly at thickprocesses that belong to reactive astrocytes.

To investigate the neuronal subcellular localizationof e-5NT, sections were probed with antibodies againste-5NT (green fluorescence) and MAP-2 (red fluores-cence) and counterstained for DAPI (blue fluorescence).In intact controls (Fig. 4A–F), fine, punctate e-5NTstaining delineates neuronal cell bodies (Fig. 4A) andMAP-2-positive primary dendrites (Fig. 4I). At 1 day

Fig. 5. Double-immunofluorescence analysis for e-5NT (green fluo-rescence) and GFAP (red fluorescence). In control sections (A–C),e-5NT immunosignal is sporadically present at rare GFAP1 struc-tures. At 4 hr (D–F) postinjury, GFAP1 astrocytes are still rare. At 1day (G–I) and 2 days (J–L) postinjury, the number of double-labeled

GFAP1/e-5NT1 astrocytes is increasing, providing an almost com-pletely overlapping signal (yellow fluorescence) at 7 days (J–L) and15 days (M–O) postinjury. Scale bar 5 20 lm. [Color figure can beviewed in the online issue, which is available at wileyonlinelibrary.com.]



postinjury (Fig. 4G–L) and afterward (data not shown),prominent loss in both e-5NT and MAP-2 immunore-activity was observed, which probably accounts for thesignificant postinjury decrease in total enzyme level at 4hr and 1 day (Fig. 1).

By means of double-immunofluorescence staining,we further investigated colocalization of e-5NT (greenfluorescence) with astrocytic markers, GFAP (Fig. 5, redfluorescence) and vimentin (Fig. 6, red fluorescence). Incontrol sections (Fig. 5A–C), as well as in sectionsobtained 4 hr postinjury (Fig. 5D–F), e-5NT signal wassporadically present in GFAP-positive structures scatteredamong neurons. Starting from day 2 postinjury (Fig. 5J–L) and afterward, the GFAP immunosignal shows a pat-tern of astrocytic reaction typical for cortical injury. At 2days postinjury, the number of double GFAP-positive/e-5NT-positive cells increases, providing an almostcompletely overlapping signal (yellow fluorescence) at 7(Fig. 5M–O) and 15 (Fig. 5P–R) days postinjury. A sim-ilar staining pattern was obtained by triple-fluorescence

staining for e-5NT (green fluorescence), vimentin (redfluorescence), and DAPI (blue fluorescence; Fig. 6). Incontrol sections (data not shown), as well as sectionsobtained 4 hr postinjury (Fig. 6A–C), vimentin-positivecells were not observed. At 2 days (Fig. 6D–F) andmore prominently at 7 days (Fig. 6G–I) postinjury,strong vimentin immunofluorescence completely over-laps with e-5NT-reactive astrocytes. At 15 days (Fig.6J–L) postinjury, vimentin immunoreactivity decreases,disappearing from astrocytic bodies and residing on finee-5NT-positive astrocytic processes.

DISCUSSION

Numerous studies document that brain injuryinduces massive release of ATP and other nucleotides inthe extracellular space (Melani et al., 2005; Franke et al.,2006b; Burnstock, 2008). The consequences of highextracellular ATP in a specific brain area are determinedmainly by the availability of P2 receptors and the activity

Figure 5. (Continued)



of NTPDases, which account for rapid removal of ATP.However, NTPDases control not only the level ofextracellular ATP but also the level of ADP and, to-gether with e-5NT, the level of adenosine, which are

native agonist for distinct P2Y and P1 receptors, respec-tively.

The focus in our work has been on the e-5NTexpression in the perilesioned area after CSI in rats. Pre-

Fig. 6. Triple-fluorescence staining for e-5NT (green fluorescence),vimentin (red fluorescence), and DAPI (blue fluorescence). At 4 hr post-injury (A–C), only e-5NT1 cells are seen. At 2 days (D–F) and moreprominently at 7 days (G–I) postinjury, vimentin and e-5NT immuno-

signals completely overlap. At 14 days (J–L) postinjury, vimentin immu-noreactivity dissapears from astrocytic bodies and resides only in e-5NT1 astrocytic processes. Scale bar 5 20 lm. [Color figure can beviewed in the online issue, which is available at wileyonlinelibrary.com.]



viously, we reported biphasic changes in e-5NT activityafter CSI, beginning with an acute decrease, followed bya significant up-regulation persisting 2 weeks postinjury(Nedeljkovic et al.,, 2006). Similar time-course changesin e-5NT protein expression are shown in the presentstudy by immunoblot and immunohistochemical analy-ses. These results raised two major questions: what arethe mechanisms for down and up-regulation of the pro-tein levels and the enzyme activity after CSI, and whatare physiological consequences of such biphasic changes?

It was recently shown that inhibition of e-5NTnot only results in accumulation of AMP but also buildsup extracellular ATP concentration after in vivo ische-mia (Melani et al., 2005). Thus, an early decrease ine-5NT protein level and the enzyme activity suggestthat CSI, because of the active release of nucleotides andthe lack of efficient metabolism, results in the accumula-tion of purine nucleotides and facilitation of P2X andP2Y receptor signaling in the acute phase of injury. Incontrast, prominent up-regulation of e-5NT protein andincrease in the activity of the whole enzyme chain ofectonucleotidases later after injury (Braun et al., 1998;Nedeljkovic et al., 2006) imply efficient degradation ofATP and accumulation of adenosine in the perilesionedarea. Because all neuronal and nonneuronal cell typesexpress variable P2 and P1 receptors (Abbracchio andBurnstock, 1998; Ochiishi et al., 1999; Cunha, 2005;Verkhratsky et al., 2009), it is reasonable to concludethat dynamic changes in e-5NT expression alterextracellular levels of nucleotides and the ATP/adeno-sine ratio, directly influencing cellular and biochemicalresponses to CSI.

The most interesting finding in the present study isthe prominent loss of e-5NT immunoreactivity fromneurons in the perilesioned area soon after CSI. Namely,under physiological conditions, e-5NT is abundant onneuronal cell bodies and apical dendrites and on fineastrocytic processes scattered among neurons. Early afterCSI, there was a significant decrease in e-5NT labeling,first from the apical dendrites and later from neuronalcell bodies. Because CSI causes cell death, this couldaccount for a decrease in e-5NT levels immediately(4 hr) after injury. We observed certain level of neuronalcell loss in the perilesioned area, but cell loss alone isunlikely to account for the total decrease in neuronale-5NT immunoreactivity.

Several studies indicate that the e-5NT gene pro-moter contains cAMP response element (CRE; Hansenet al., 1995) and that CREB binding is critical for basalexpression of the enzyme (Narravula et al., 2000).Assessment of genomic alterations using high-densitymicroarray analysis (Matzilevich et al., 2002) and single-neuron mRNA profiling (O’Dell et al., 2000) after trau-matic brain injury revealed significant decrease in CREBgene expression lasting for 12 hr postinjury. The pro-moter region of e-5NT gene also has consensus sites forat least two additional transcription factors, GATA-1 andGATA-2 (Heinemeyer et al., 1998), that have beenimplicated in repression of genes in hypoxia (De Maria

et al., 1999; Tarumoto et al., 2000). Thus, the decreasein e-5NT protein levels was most probably caused bydown-regulation of e-5NT mRNA expression resultingfrom changes in the transcription factor levels. However,although this can explain the lack of recovery of theenzyme expression in neurons in CSI animals, it cannotexplain the rapid decrease in the enzyme protein level assoon as 4 hr after the injury.

The diffuse staining pattern of e-5NT fluorescenceimmediately after the injury prompted us to suggestanother mechanism accounting for the rapid loss of theenzyme in the plasma membrane of neurons. e-5NT is aglycosylphosphatidylinositol-anchored enzyme that couldbe cleaved off the membrane by phosphatidylinositol-specific phospholipase C (Lehto and Sharom, 1998,2002). Although phosphatidylinositol-specific phospholi-pase C-mediated cleavage of e-5NT was indicated invascular endothelial cells (Kalsi et al., 2002), it remainsto be seen whether the same mechanism could be re-sponsible for the decrease in e-5NT surface expression at4 hr after CSI.

The sustained up-regulation of enzyme levels isanother major finding shown in this study. The recoveryof expression of the enzyme in neurons was not the reasonfor the increase in total protein level in the injured area,but it reflects up-regulation of expression of e-5NT inastrocytes and/or movement of e-5NT-positive cells intothe damaged region. Among the various cell types in theCNS, astrocytes are the first target of nucleotides and theirtrophic actions. In this regard, it is well known that highlevels of ATP early after CSI up-regulate P2X and P2Yreceptors (Neary and Kang, 2005; Franke et al., 2006b)and activate ERK (Neary et al., 2003), Akt (Neary et al.,2005) and glycogen synthase kinase-3b in astrocytes(Neary and Kang, 2006), leading to their rapid prolifera-tion, stellation, and migration to the site of injury. In thepresent CSI model, starting from day 2 postinjury, astro-cytes reacted in a typical hypertrophic and hyperplasticmanner, with an increase in immunoreactivity for GFAPand vimentin. However, in additition to these well-described morphological alterations, the functional conse-quences of astrocyte reactivity are less well characterizedand seem to depend on the molecular pathways involved.The present study shows that prolonged postinjury sur-vival time results in the massive migration of reactiveastrocytes into the perilesioned area and that temporal andspatial development of GFAP immunoreactivity com-pletely corresponds to immunoreactivity for e-5NT.Although we have not addressed this issue, it should benoted that ATP also induces microglial chemotaxis ininjured tissue (Honda et al., 2001) and mediates processretraction in activated microglia (Davalos et al., 2005;Gyoneva et al., 2009). Thus, an increase in the amountand/or expression of e-5NT in reactive astrocytes (presentstudy) and NTPDase by activated microglia (Braun et al.,1998; Nedeljkovic et al., 2006) may reinforce the extrac-ellular purine salvage pathway and enhance adenosineneuroprotective and neurotrophic functions (Cunha,2005) that could be relevant for neuronal survival.



In summary, the main goal of the present studywas to elucidate the temporal and spatial (cellular)pattern of e-5NT expression in the perilesioned area af-ter CSI in rats. Our results indicate the presence ofenzyme in the plasma membranes surrounding neuronalsomata and apical dendrites and less frequently in plasmamembrane of astrocytes in control tissue. After CSI, weobserved biphasic changes in the total enzyme levels ininjured region, an initial decrease that lasted for 2 days,followed by a gradual increase in the enzyme levelsduring the 15-day period. We further found that the lossof the enzyme from neuronal membranes accounted fora rapid decrease in the total protein level, whereas anincrease in e-5NT-positive astrocytes in the injuredbrain area accounted for the increase in total proteinlevels.

ACKNOWLEDGMENTS

We thank Dr. Markus Kipp, University Aachen,for useful comments on the results. The authors aregrateful to Dr. Stanko Stojilkovic, NIH, for critical read-ing of the manuscript, intellectual support, and advice.

REFERENCES

Abbracchio MP, Burnstock G. 1998. Purinergic signalling: pathophysio-

logical roles. Jpn J Pharmacol 78:113–145.

Amadio S, Cavaliere F, Murra B, Sancesario G, Bernardi G, Burnstock

G, Volonte C. 2002. P2 receptor modulation and cytotoxic function in

cultured CNS neurones. Neurophyrmacology 42:489–501.

Bjelobaba I, Lavrnja I, Parabucki A, Stojkov D, Stojiljkovic M, Pekovic

S, Nedeljkovic N. 2010. The cortical stab injury induces beading of

fibers expressing ecto-nucleoside triphosphate diphosphohydrolase 3.

Neuroscience 170:107–116.

Bonan CD, Amaral OB, Rockenbach IC, Walz R, Battastini AM,

Izquierdo I, Sarkis JJ. 2000a. Altered ATP hydrolysis induced by penty-

lenetetrazol kindling in rat brain synaptosomes. Neurochem Res

25:775–779.

Bonan CD, Walz R, Pereira GS, Worm PV, Battastini AM, Cavalheiro

EA, Izquierdo I, Sarkis JJ. 2000b. Changes in synaptosomal ectonucleo-

tidase activities in two rat models of temporal lobe epilepsy. Epilepsy

Res 39:229–238.

Braun N, Zhu Y, Krieglstein J, Culmsee C, Zimmermann H. 1998. Up-

regulation of the enzyme chain hydrolyzing extracellular ATP after

transient forebrain ischemia in the rat. J Neurosci 18:4891–4900.

Burnstock G. 2008. Purinergic signalling and disorders of the central

nervous system. Nat Rev Drug Discov 7:575–590.

Chen S, Pickard JD, Harris NG. 2003. Time course of cellular pathology

after controlled cortical impact injury. Exp Neurol 182:87–102.

Cunha RA. 2005. Neuroprotection by adenosine in the brain: from A1

receptor activation to A2A receptor blockade. Purinergic Signal 1:111–

134.

Davalos D, Grutzendler J, Yang G, Kim JV, Zuo Y, Jung S, Littman

DR, Dustin ML, Gan WB. 2005. ATP mediates rapid microglial

response to local brain injury in vivo. Nat Neurosci 8:752–758.

De Maria R, Zeuner A, Eramo A, Domenichelli C, Bonci D, Grignani

F, Srinivasula SM, Alnemri ES, Testa U, Peschle C. 1999. Negative

regulation of erythropoiesis by caspase-mediated cleavage of GATA-1.

Nature 401:489–493.

Di Virgilio F, Ceruti S, Bramanti P, Abbracchio MP. 2009. Purinergic

signalling in inflammation of the central nervous system. Trends Neu-

rosci 32:79–87.

Edwards FA. 1996. Features of P2X receptor-mediated synapses in the

rat brain: why doesn’t ATP kill the postsynaptic cell? Ciba Found

Symp 198:278–286; discussion 286–279.

Franke H, Grummich B, Hartig W, Grosche J, Regenthal R, Edwards

RH, Illes P, Krugel U. 2006a. Changes in purinergic signaling after

cerebral injury—involvement of glutamatergic mechanisms? Int J Dev

Neurosci 24:123–132.

Franke H, Krugel U, Illes P. 2006b. P2 receptors and neuronal injury.

Pflugers Arch 452:622–644.

Gray EG, Whittaker VP. 1962. The isolation of nerve endings from

brain: an electron-microscopic study of cell fragments derived by ho-

mogenization and centrifugation. J Anat 96:79–88.

Gyoneva S, Orr AG, Traynelis SF. 2009. Differential regulation of

microglial motility by ATP/ADP and adenosine. Parkinsonism Rel Dis-

ord 15(Suppl 3):S195–S199.

Hansen KR, Resta R, Webb CF, Thompson LF. 1995. Isolation and

characterization of the promoter of the human 50-nucleotidase (CD73)-

encoding gene. Gene 167:307–312.

Haynes SE, Hollopeter G, Yang G, Kurpius D, Dailey ME, Gan WB,

Julius D. 2006. The P2Y12 receptor regulates microglial activation by

extracellular nucleotides. Nat Neurosci 9:1512–1519.

Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE, Kel

OV, Ignatieva EV, Ananko EA, Podkolodnaya OA, Kolpakov FA,

Podkolodny NL, Kolchanov NA. 1998. Databases on transcriptional

regulation: TRANSFAC, TRRD and COMPEL. Nucleic Acids Res

26:362–367.

Honda S, Sasaki Y, Ohsawa K, Imai Y, Nakamura Y, Inoue K, Kohsaka

S. 2001. Extracellular ATP or ADP induce chemotaxis of cultured

microglia through Gi/o-coupled P2Y receptors. J Neurosci 21:1975–

1982.

Inoue K, Koizumi S, Nakazawa K. 1995. Glutamate-evoked release of

adenosine 50-triphosphate causing an increase in intracellular calcium in

hippocampal neurons. Neuroreport 6:437–440.

James S, Richardson PJ. 1993. Production of adenosine from extracellular

ATP at the striatal cholinergic synapse. J Neurochem 60:219–227.

Kalsi K, Lawson C, Dominguez M, Taylor P, Yacoub MH, Smolenski

RT. 2002. Regulation of ecto-50-nucleotidase by TNF-alpha in human

endothelial cells. Mol Cell Biochem 232:113–119.

Koizumi S, Shigemoto-Mogami Y, Nasu-Tada K, Shinozaki Y, Ohsawa

K, Tsuda M, Joshi BV, Jacobson KA, Kohsaka S, Inoue K. 2007. UDP

acting at P2Y6 receptors is a mediator of microglial phagocytosis. Na-

ture 446:1091–1095.

Laemmli UK. 1970. Cleavage of structural proteins during the assembly

of the head of bacteriophage T4. Nature 227:680–685.

Lehto MT, Sharom FJ. 1998. Release of the glycosylphosphatidylinosi-

tol-anchored enzyme ecto-50-nucleotidase by phospholipase C: catalytic

activation and modulation by the lipid bilayer. Biochem J 332:101–109.

Lehto MT, Sharom FJ. 2002. PI-specific phospholipase C cleavage of a

reconstituted GPI-anchored protein: modulation by the lipid bilayer.

Biochemistry 41:1398–1408.

Markwell MA, Haas SM, Bieber LL, Tolbert NE. 1978. A modification

of the Lowry procedure to simplify protein determination in membrane

and lipoprotein samples. Anal Biochem 87:206–210.

Matzilevich DA, Rall JM, Moore AN, Grill RJ, Dash PK. 2002. High-

density microarray analysis of hippocampal gene expression following

experimental brain injury. J Neurosci Res 67:646–663.

Melani A, Turchi D, Vannucchi MG, Cipriani S, Gianfriddo M, Pedata

F. 2005. ATP extracellular concentrations are increased in the rat stria-

tum during in vivo ischemia. Neurochem Int 47:442–448.

Nagy AK, Walton NY, Treiman DM. 1997. Reduced cortical ecto-

ATPase activity in rat brains during prolonged status epilepticus induced

by sequential administration of lithium and pilocarpine. Mol Chem

Neuropathol 31:135–147.



Narravula S, Lennon PF, Mueller BU, Colgan SP. 2000. Regulation of

endothelial CD73 by adenosine: paracrine pathway for enhanced endo-

thelial barrier function. J Immunol 165:5262–5268.

Neary JT, Kang Y. 2005. Signaling from P2 nucleotide receptors to pro-

tein kinase cascades induced by CNS injury: implications for reactive

gliosis and neurodegeneration. Mol Neurobiol 31:95–103.

Neary JT, Kang Y. 2006. P2 purinergic receptors signal to glycogen syn-

thase kinase-3beta in astrocytes. J Neurosci Res 84:515–524.

Neary JT, Kang Y, Willoughby KA, Ellis EF. 2003. Activation of extrac-

ellular signal-regulated kinase by stretch-induced injury in astrocytes

involves extracellular ATP and P2 purinergic receptors. J Neurosci

23:2348–2356.

Neary JT, Kang Y, Tran M, Feld J. 2005. Traumatic injury activates pro-

tein kinase B/Akt in cultured astrocytes: role of extracellular ATP and

P2 purinergic receptors. J Neurotrauma 22:491–500.

Neary JT, Kang Y, Shi YF, Tran MD, Wanner IB. 2006. P2 receptor

signalling, proliferation of astrocytes, and expression of molecules

involved in cell–cell interactions. Novartis Found Symp 276:131–143;

discussion 143–137, 233–137, 275–181.

Nedeljkovic N, Nikezic G, Horvat A, Pekovic S, Stojiljkovic M, Marti-

novic JV. 1998. Properties of Mg21-ATPase rat brain synaptic plasma

membranes. Gen Physiol Biophys 17:3–13.

Nedeljkovic N, Banjac A, Horvat A, Stojiljkovic M, Nikezic G. 2005.

Developmental profile of NTPDase activity in synaptic plasma mem-

branes isolated from rat cerebral cortex. Int J Dev Neurosci 23:45–51.

Nedeljkovic N, Bjelobaba I, Subasic S, Lavrnja I, Pekovic S, Stojkov D,

Vjestica A, Rakic L, Stojiljkovic M. 2006. Up-regulation of ectonu-

cleotidase activity after cortical stab injury in rats. Cell Biol Int 30:541–

546.

Nedeljkovic N, Bjelobaba I, Lavrnja I, Stojkov D, Pekovic S, Rakic L,

Stojiljkovic M. 2008. Early temporal changes in ecto-nucleotidase activ-

ity after cortical stab injury in rat. Neurochem Res 33:873–879.

O’Dell DM, Raghupathi R, Crino PB, Eberwine JH, McIntosh TK.

2000. Traumatic brain injury alters the molecular fingerprint of

TUNEL-positive cortical neurons in vivo: a single-cell analysis. J Neu-

rosci 20:4821–4828.

O’Regan MH, Simpson RE, Perkins LM, Phillis JW. 1992. Adenosine

receptor agonists inhibit the release of gamma-aminobutyric acid

(GABA) from the ischemic rat cerebral cortex. Brain Res 582:22–26.

Ochiishi T, Chen L, Yukawa A, Saitoh Y, Sekino Y, Arai T, Nakata H,

Miyamoto H. 1999. Cellular localization of adenosine A1 receptors in

rat forebrain: immunohistochemical analysis using adenosine A1 recep-

tor-specific monoclonal antibody. J Comp Neurol 411:301–316.

Ohsawa K, Irino Y, Nakamura Y, Akazawa C, Inoue K, Kohsaka S.

2007. Involvement of P2X4 and P2Y12 receptors in ATP-induced

microglial chemotaxis. Glia 55:604–616.

Ridet JL, Malhotra SK, Privat A, Gage FH. 1997. Reactive astrocytes:

cellular and molecular cues to biological function. Trends Neurosci

20:570–577.

Sofroniew MV. 2005. Reactive astrocytes in neural repair and protection.

Neuroscientist 11:400–407.

Stone TW, Forrest CM, Mackay GM, Stoy N, Darlington LG. 2007.

Tryptophan, adenosine, neurodegeneration and neuroprotection. Metab

Brain Dis 22:337–352.

Takenouchi T, Sugama S, Iwamaru Y, Hashimoto M, Kitani H. 2009.

Modulation of the ATP-induced release and processing of IL-1beta in

microglial cells. Crit Rev Immunol 29:335–345.

Tarumoto T, Imagawa S, Ohmine K, Nagai T, Higuchi M, Imai N,

Suzuki N, Yamamoto M, Ozawa K. 2000. N(G)-monomethyl-

L-arginine inhibits erythropoietin gene expression by stimulating

GATA-2. Blood 96:1716–1722.

Verkhratsky A, Krishtal OA, Burnstock G. 2009. Purinoceptors on neu-

roglia. Mol Neurobiol 39:190–208.

Villa RF, Gorini A, Hoyer S. 2002. ATPases of synaptic plasma mem-

branes from hippocampus after ischemia and recovery during ageing.

Neurochem Res 27:861–870.

Wang M, Kong Q, Gonzalez FA, Sun G, Erb L, Seye C, Weisman GA.

2005. P2Y nucleotide receptor interaction with alpha integrin mediates

astrocyte migration. J Neurochem 95:630–640.

Xiang Z, Chen M, Ping J, Dunn P, Lv J, Jiao B, Burnstock G. 2006.

Microglial morphology and its transformation after challenge by extrac-

ellular ATP in vitro. J Neurosci Res 83:91–101.

Yegutkin GG. 2008. Nucleotide- and nucleoside-converting ectoen-

zymes: Important modulators of purinergic signalling cascade. Biochim

Biophys Acta 1783:673–694.

Zimmermann H. 2000. Extracellular metabolism of ATP and other

nucleotides. Naunyn Schmiedebergs Arch Pharmacol 362:299–309.



J Neurosci Res

Documents

Transcript of J Neurosci Res