Distribution and expression of A1 adenosine receptors, adenosine deaminase and adenosine deaminase-binding protein (CD26) in goldfish brain

Neurochemistry International 42 (2003) 455–464

Distribution and expression of A1 adenosine receptors, adenosine deaminase and adenosine deaminase-binding protein (CD26) in goldfish brain A. Beraudi a , U. Traversa b , L. Villani a , Y. Sekino c , J.I. Nagy d , A. Poli a,∗ a Department of Evolutionistic and Experimental Biology, University of Bologna, Via Selmi 3, 40126 Bologna, Italy Department of Biomedical Sciences, B.R.A.I.N. Centre for Neuroscience, University of Trieste, Via L. Giorgieri 7, 34127 Trieste, Italy Department of Neurobiology and Behaviour, Gunma University School of Medicine, 3-39-22 Showa-machi, Maebashi 371-8511, Japan d Department of Physiology, University of Manitoba, Basic Medical Sciences Building, RM 408,730 William Avenue, Winnipeg, Man., Canada R3E 3J7

b c

Received 3 June 2002; received in revised form 16 September 2002; accepted 2 October 2002

Abstract The expression patterns of adenosine A1 receptors (A1 Rs), adenosine deaminase (ADA) and ADA binding protein (CD26) were studied in goldfish brain using mammalian monoclonal antibody against A1 R and polyclonal antibodies against ADA and CD26. Western blot analysis revealed the presence of a band of 35 kDa for A1 R in membrane preparations and a band of 43 kDa for ADA in both cytosol and membranes. Immunohistochemistry on goldfish brain slices showed that A1 receptors were present in several neuronal cell bodies diffused in the telencephalon, cerebellum, optic tectum. In the rhombencephalon, large and medium sized neurons of the raphe nucleus showed a strong immunopositivity. A1 R immunoreactivity was also present in the glial cells of the rhombencephalon and optic tectum. An analogous distribution was observed for ADA immunoreactivity. Tests for the presence of CD26 gave positive labelling in several populations of neurons in the rhombencephalon as well as in the radial glia of optic tectum, where immunostaining for ADA and A1 R was observed. In goldfish astrocyte cultures the immunohistochemical staining of A1 R, ADA and CD26, performed on the same cell population, displayed a complete overlapping distribution of the three antibodies. The parallel immunopositivity, at least in some discrete neuronal areas, for A1 Rs, ADA and CD26 led us to hypothesize that a co-localization among A1 R, ecto-ADA and CD26 also exists in the neurons of goldfish since it has been established to exist in the neurons of mammals. Moreover, we have demonstrated for the first time, that A1 R, ecto-ADA and CD26 co-localization is present on the astroglial component of the goldfish brain. This raises the possibility that a similar situation is also shown in the glia of the mammalian brain. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: A1 receptor; Adenosine deaminase; CD26 protein; Goldfish; Central nervous system; Astrocyte culture; Immunohistochemistry; Western blotting

1. Introduction Adenosine A1 receptors (AR1 s) have been shown to have a broad phylogenetic distribution (Siebenaller and Murray, 1986). In particular, our studies performed in teleost brain have shown that A1 Rs appeared pharmacologically and biochemically similar to that found in mammals, and their activation seemed to be coupled to the same intracellular

Abbreviations: A1 R, adenosine A1 receptor; ADA, adenosine deaminase; CD26, dipeptidylpeptidase IV; CNS, central nervous system; EDTA, ethylenediamine tetraacetic acid; HPR, horseradish peroxidase; ECL, enhanced chemioluminescence; PBS, phosphate buffer solution; FCS, fetal calf serum; GFAP, glial fibrillary acidic protein ∗ Corresponding author. Tel: +39-51-2094135; fax: +39-51-251208. E-mail address: [email protected] (A. Poli).

signalling pathways described in mammalian brain (Lucchi et al., 1992, 1994). Studies performed in mammals show that the steady-state concentration of adenosine is maintained by the activities of a small group of enzymes, including adenosine deaminase, which catalyses the conversion of the adenosine into its inactive metabolite inosine. Immunohistochemical localization of ADA has been reported in discrete neuronal populations in the brain of different mammals (Yamamoto et al., 1987). Besides the main cytosolic localization of the enzyme, adenosine deaminase has been found on the surface of many cell types, including brain synaptosomes (Franco et al., 1986, 1997). Thus, ADA maintains control of the extracellular adenosine availability, suggesting that the suppressive tonic control of neuronal activity mediated by A1 R is governed in part by ecto-ADA. A co-localization of A1 Rs and ecto-ADA

0197-0186/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 9 7 - 0 1 8 6 ( 0 2 ) 0 0 1 5 7 - 2

456

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

has been shown in neuronal and non-neuronal cells (Ciruela et al., 1996; Franco et al., 1998; Ruiz et al., 2000). A1 R can act as a cell surface molecule which is able to interact with ecto-ADA. The interaction between the two proteins is the functional basis of the extra-enzymatic role of ecto-ADA in modulating ligand-induced signalling, desensitization and internalization of A1 Rs (Ciruela et al., 1996; Saura et al., 1998; Gines et al., 2001). The cell surface CD26 protein has been shown to be another protein anchoring ecto-ADA found in different cells (Kameoka et al., 1993; Dong et al., 1996). The ADA/CD26 complex has been suggested to trigger signalling pathways in lymphocytes (Martin et al., 1995; Franco et al., 1998). As cells in the nervous system express ecto-ADA, CD26, and A1 R, ecto-ADA seems to be able to transmit signals when interacting with either A1 R or CD26 or both, thus acting as a co-stimulatory molecule in the regulation of CNS activity (Franco et al., 1997, 1998). Since the adenosine receptor system in the brain of lower vertebrates shares several functional homologies with that in the mammalian brain, this research aims to verify if the distribution pattern of A1 Rs, ADA and CD26 proteins observed in mammals is also present in the brain of the teleost Carassius auratus (goldfish). The immunohistochemical studies were performed on brain sections. Moreover, in order to better discriminate between neuronal and glial cells, the immunoreactivity was also performed in astrocyte cultures from goldfish brain.

and filtered through activated charcoal filters. Fish were fed once daily with a commercial fish pellet and kept under a 12 h:12 h light–dark photoperiod. 2.3. Membrane and cytosol preparation For membrane preparation, goldfish brains removed following decapitation were dissected on ice and homogenized in 10 ml/g of ice-cold buffer (10% (w/v) sucrose in 10 mM phosphate buffer solution, pH 7.4) by using Ultra Turrax (Ika-Werk, Germany). The homogenate was centrifuged at 1000 × g for 10 min at 4 ◦ C. The supernatant was kept and centrifuged al 100,000 × g for 40 min at 4 ◦ C. The supernatant was discarded and the pellet was resuspended in 2.5 ml/g of ice-cold buffer (50 mM Tris–HCl and 1 mM EDTA pH 7.4, containing protease inhibitors cocktail). For the cytosol preparation, goldfish brain were removed following decapitation, dissected on ice and homogenized in 2.5 ml/g of ice-cold buffer (50 mM Tris–HCl and 1 mM EDTA, pH 7.4, containing protease inhibitors cocktail) with Ultra Turrax (Ika-Werk, Germany). The homogenate was briefly sonicated and centrifuged at 100,000× g for 30 min at 4 ◦ C. The supernatant was collected and the pellet discarded. Aliquots of both preparations were stored at −80 ◦ C until Western blot analysis. Protein determination was made by Lowry et al. (1951) method. 2.4. Western blot

2. Experimental procedure 2.1. Materials Antibody against purified calf intestinal ADA was generated in rabbit as previously described (Daddona and Kelley, 1977; Nagy et al., 1996); its concentration was 0.5 mg/ml. A rabbit antibody against human kidney ADA-CP (CD26) was prepared as previously described (final concentration 2 mg/ml) and shown to be highly specific for ADA-CP in neural and peripheral tissues (Schrader et al., 1987). The anti-A1 R monoclonal antibody used in this study was a gift of Dr. H. Nakata (Nakata, 1993); its concentration was 4.4 mg/ml. The antibody developed in rabbit against GFAP was purchased from DAKO (4.1 mg/ml). ECL detection reagent, nitrocellulose membrane, streptavidin horseradish peroxidase and secondary antibodies anti-mouse and anti-rabbit used in immunohistochemistry assay were purchased from Amersham. All other chemicals were supplied by SIGMA.

Western blots of goldfish membranes and cytosol preparations were probed with a monoclonal antibody for A1 R and a polyclonal antibody for ADA. The Western blots were compared to those performed using rat brain homogenate. The 260 ␮g protein samples were suspended into an equal volume of Laemmli’s sample buffer. The samples were subsequently applied to 10% sodium dodecyl sulfate polyacriylamide gels and the separated proteins were transferred to a nitrocellulose membrane. The membranes were blocked by 1 h incubation in PBS containing 0.1% Tween-20 and 5% no fat skim milk and then incubated overnight with the primary antibody:anti-mouse A1 R (not diluted) or anti-rabbit ADA 1:5000. The following day, the membranes were washed in PBS with Tween-20 and 5% no fat skim milk and then incubated for 90 min with the secondary HPR conjugated antibody diluted 1:2000. After prolonged washing in PBS, containing 0.1% Tween-20 and 5% no fat skim milk, the membranes were incubated in equal volumes of ECL detection reagent for 1 min each. The reagent was drained off, and membranes were placed in contact with a film, which was developed by chemiluminescence.

2.2. Animals

2.5. Tissue preparation

Goldfish (Carassius auratus) obtained from a local breeder, were maintained at 22 ◦ C in thermostated-glass aquaria filled with unchlorinated water constantly aerated

Goldfish (12–14 cm length) were deeply anaesthetised with 3-aminobenzoic acid ethyl-ester in proportion with body weight (20 mg/kg) and perfused transcardially with

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

cold PBS 0.1 M containing heparin (0.5 mg/ml) and 1% sodium nitrite. This was followed by perfusion with cold PBS containing 4% paraformaldehyde. Brains were removed, post fixed in the perfusion fixative for 2–6 h at 4 ◦ C and then cryoprotected in PBS 0.1 M containing 25% sucrose for 48 h. Using a cryostat, 20 ␮m brain sections were cut. 2.6. Astrocytes cultures preparation Astrocyte cultures were prepared as previously described (Sivron Jeserich et al., 1992). Four goldfish (3 cm length) brains were used for each culture. After removal of surrounding tissue and blood, the brains were washed twice with L-15 medium. The brains were fragmented pipetting several times, and incubated for 40 min at room temperature with 2 ml trypsin (15,000 U/ml). A 2 ml solution, containing 5200 U/ml soybean trypsin inhibitor, 74 U/ml DNase I, 3 mg/ml bovine serum albumin was then added. The 1 min later the preparation was centrifuged (150 × g for 5 min at 5 ◦ C). The supernatant was removed and 1 ml of L-15 medium was added to the pellet. The pellet was dissociated by pipetting several times. Following the addition of 6 ml of L-15, the cell suspension was allowed to stand for 10 min to let any undissociated pieces settle. To this suspension, 9 ml L-15 medium plus 1% fetal calf serum (FCS) was added. The cell suspension was pipetted until homogenous, and then seeded in 5 ml aliquots on Tissue Culture glass chamber slides. The chambers were precoated overnight with poly-l-lysine (20 ␮g/ml). Following 24 h the medium was replaced by L-15 supplemented with 5% FCS plus 0.1 mg/ml gentamicin. The cells were grown at 24 ◦ C for 15 days.

457

CD26. Cultures were washed one time with L-15 medium and fixed for 15 min at 4 ◦ C with 4% (w/v) paraformaldehyde in phosphate buffer, pH 7.4. After a permeabilization for three times in PBS with 0.1% Triton X-100 (PBS-T), cultured cells were incubated for 30 min with PBS-T containing 1% bovine serum albumin (BSA). After two washing in PBS-T cultures were incubated for 2 h at room temperature with two primary antibodies developed in different animals using the same dilutions adopted for tissue sections. The cells were then washed two times in PBS-T and labelled for 90 min at room temperature with the appropriate secondary antibodies for the primary ones. The secondary antibodies, conjugated with two different fluorescence labelling were diluted in PBS-T, 1:400 for anti-rabbit rhodamine conjugated, and 1:40 for the anti-mouse fluorescent conjugated. After a washing in PBS the chamber slides were covered by a coverslip mounted with antifading and observed by fluorescence microscope equipped with appropriate filters for viewing fluorescein and rhodamine. The co-localizations for ADA with GFAP and ADA with CD26 were performed with sequential technique: firstly for one molecule and then for the second one, restarting from the incubation with the primary antibody. The experiments reported here were performed according to the Italian law on protection of laboratory animals, with the approval of a local bioethical committee and under the supervision of a veterinary commission for control of animal care and comfort.

2.7. Immunohistochemistry for tissue sections After 15 min preincubation in H2 O2 0.5% in PBS, the brain sections were incubated for 48 h with antibodies anti-ADA and anti-CD26 diluted 1:500 in PBS 0.1 M containing 0.3% Triton X-100 (PBS-T) and 1% BSA, or with antibody for A1 R (undiluted). Sections were then washed three times (10ute each) in PBS-T and incubated for 2 h at room temperature with biotinylated secondary antibodies anti-rabbit for ADA and CD26 (1:30) and anti-mouse for A1 R (1:200). The sections were washed for three times in PBS-T and incubated for 90 min at room temperature with streptavidin horseradish peroxidase diluted 1:100 in PBS-T. The sections were then washed for 30 min in PBS-T followed by a 30 min washing in 50 mM Tris–HCl pH 7.4, and then were reacted for 5–10 min with 3,3 -diaminobenzidine (DAB). After a 20 min wash in 50 mM Tris–HCl buffer, the sections were dehydrated and coverslips were added. 2.8. Immunohistochemistry for cell culture Simultaneous technique was performed for the co-localization of A1 R with GFAP, A1 R with ADA and A1 R with

Fig. 1. Western blots detection of A1 R and ADA in rat cortex homogenate and in membrane and cytosol preparations from goldfish brain. Western blot of anti-mouse A1 R antibody using rat brain cortex homogenate (Rat), goldfish brain membranes (GFm) and goldfish brain cytosol preparation (GFc). The molecular weight (35 kDa) of the receptor in rat cortex corresponds to that observed in goldfish brain membranes. In goldfish cytosol preparation the presence of the receptor protein is very low. The antibody anti-rabbit against ADA reacts with a single protein of 43 kDa in rat, GFm and GFc samples.

458

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

3. Results 3.1. Western blot analysis Western blots were performed on membrane and cytosol samples obtained from the whole brain of goldfish. A sample from rat cortex homogenate was used as a control. Immunoblot for A1 receptor showed that the anti-mouse

antibody recognised a solitary band in goldfish membranes corresponding to that of mammals (35 kDa) (Fig. 1). In goldfish cytosol preparation, a very weak signal for A1 R appeared. The lower band shown in rat sample is probably due to the deglycosilate component of the receptor present in the whole brain homogenate (Ochiishi et al., 1999). The polyclonal anti-rabbit antibody for ADA showed a band in both membrane and cytosol goldfish preparations with an

Fig. 2. Immunoperoxidase staining of goldfish brain for A1 receptor. Several labelled neurons are seen in telencephalon (A), in raphe nucleus of rhomboencephalon (arrows) (B), in the granular layer (gl) of cerebellum (C) and in the stratum periventricularis (spv) of optic tectum, while scattered neurons are stained in the stratum griseum centralis (sgc) (D), glial distribution of the receptor in the stratum marginalis (sm) of the optic tectum (E), and in the basal part of rhombencephalon (F). ml: molecular layer; so: stratum opticum; bars, A: 100 ␮m; B: 50 ␮m; C, D: 150 ␮m; E: 200 ␮m.

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

459

apparent 43 kDa molecular weight corresponding to that observed in rat homogenate (Fig. 1).

proteins in goldfish brain sections and in culture enriched with astrocytes.

3.2. Immunohistochemical assays

3.2.1. Distribution of adenosine A1 receptor immunoreactivity A positive staining for A1 R was present in several neuronal structures in discrete brain areas as shown in sagittal sections of the goldfish brain. Neuronal cell bodies with

Immunohistochemical assays with monoclonal anti-mouse antibody against A1 R and polyclonal anti-rabbit ADA and CD26 were performed to localize the distribution of these

Fig. 3. Immunoperoxidase distribution for adenosine deaminase of goldfish brain. Immunopositive neurons are localised in telencephalon (A), in the raphe nucleus of the rhomboencephalon (arrows) (B), in the granular layer (gl) of the cerebellum (C) and in the optic tectum (D), glial immunolocalization of ADA in the stratum marginalis (sm) of the optic tectum (E), and in the basal part of rhombencephalon (F). ml: molecular layer; sgc: stratum griseum centrale; so: stratum opticum; spv: stratum periventriculare; bars, A: 100 ␮m; B: 50 ␮m; C, D: 150 ␮m; E: 200 ␮m.

460

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

a strong positivity appeared scattered in the telencephalic hemisphere (Fig. 2A). As well, in the rhombencephalon large and medium sized neurons of the raphe nucleus appeared strongly labelled (Fig. 2B). A marked immunopositivity was present: in the granular layer of the cerebellum (Fig. 2C), in the cell bodies of the stratum periventricularis, as well as in several neurons of the stratum griseum centralis of the optic tectum (Fig. 2D). A1 R antibody stained also glial structures. The immunopositivity of the A1 R was mainly present in the radially oriented processes of glial cells spanning into the stratum marginalis of the optic tectum (Fig. 2E) and into the glial component of the romboencephalon (Fig. 2F). 3.2.2. Distribution of adenosine deaminase immunoreactivity As shown in Fig. 3, the ADA immunoreactivity appeared distributed in the same areas where an evident immunopositivity for A1 R was present. Several telencephalic neurons appeared positively labelled (Fig. 3A), whereas a more intense staining of the cell body was shown in the neurons of raphe nucleus (Fig. 3B). Granular layer of cerebellum exhibited an evident immunopositivity (Fig. 3C). In optic tectum sections, a large cell population of the stratum periventricularis was strongly marked, whereas few neurons in the stratum griseum centralis were stained (Fig. 3D). The ADA immunostaining was seen in the radial glia of the stratum marginalis in the optic tectum by the ADA antibody (Fig. 3E), as well as in several processes of the glial cells in the rhomboencephalon (Fig. 3F). 3.2.3. Immunohistochemical localization of CD26 Since we revealed a band reacting for ADA by immunoblot analysis in goldfish brain membranes, we tested for the presence of CD26 in brain slices as a reasonable candidate for the mediation of this interaction. (Kameoka et al., 1993; Dong et al., 1996). As shown in Fig. 4A and B, the antibody against the CD26 protein marked several neurons in rhomboencephalon as well as, in the optic tectum, the radial glia of the stratum marginalis where the immunostaining for A1 Rs and adenosine deaminase were also shown. 3.2.4. Localization of A1 Rs and ADA in astrocyte cultures and their co-localization Immunocytochemical assays were performed to characterize the expression of A1 Rs in primary goldfish astrocyte cultures. The glial cells, that proliferated from a small number of precursors, were identified with the staining for GFAP, a specific marker for astrocytes. Among the GFAP-positive cells, the predominant type displayed a flattened morphology with two or more cytoplasmic processes. A high expression level of A1 Rs is evident in GFAP-positive cells, both in the cell bodies and in their processes (Fig. 5A and B). The astrocytes also expressed ADA immunoreactivity, although the staining was mainly localized in the cell

Fig. 4. Immunoperoxidase reaction for CD26 in goldfish brain sections. Several neurons in the rhomboencephalon (A) show a strong staining. (B) The radial glia in the stratum marginalis of the optic tectum. Bars, A: 60 ␮m; B: 180 ␮m.

bodies and appeared less evident in the cytoplasmic processes (Fig. 5C and D). Experiments performed with the same cells immunolabelled with antibodies against A1 Rs and against ADA showed a high degree of co-localization of the two proteins (Fig. 5E and F). 3.2.5. Localization of CD26 in astrocyte cultures and co-localization with A1 Rs and ADA A high expression level of CD26 protein is evident in both astrocyte bodies and their processes. The immunocytochemical staining of CD26 and A1 R demonstrated a complete overlap in the distribution of both antibodies. All cells showed a similar pattern of labelling in the cell bodies and in the cytoplasmic processes (Fig. 6A and B). Also ADA and CD26 appeared co-localized in all astrocytes (Fig. 6C and D).

4. Discussion Our results demonstrate for the first time that neurons as well as glial cells in different regions of the goldfish brain express A1 adenosine receptors, adenosine deaminase and CD26 proteins. All of which show several homologies with mammalian brain. Moreover, the co-localizations of A1 R,

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

461

Fig. 5. Immunofluorescence labelling for A1 R, ADA and GFAP in astrocytic cell culture preparations. Co-localization of immunofluorescence labelling of A1 R and GFAP (A, B), ADA and GFAP (C, D) and A1 R and ADA (E, F) in cultures enriched with astrocytes. Bars, A, C: 15 ␮m; E: 25 ␮m.

ADA and CD26 are shown on astrocyte cultures obtained from goldfish brain.

4.1. Adenosine A1 receptors Using a monoclonal anti-adenosine A1 receptor antibody, we have shown that this receptor is distributed in several areas of the goldfish brain. Strong immunostaining showed discrete neuronal cell bodies in the raphe nucleus of rhombencephalon and in the telencephalon. An intense

immunopositivity appeared also in the cerebellum and in some layers of the optic tectum. The immunohistochemical detection of A1 adenosine receptors confirms our previous findings obtained in membrane preparations from different regions of goldfish brain by radioligand binding assays (Lucchi et al., 1992, 1994). Although, most goldfish brain areas are not homologous to those present in mammals, some analogies exist in the distribution of A1 Rs in the goldfish brain with that shown in the mammalian brain. The raphe nucleus of the romboencephalon in fish is considered homologous to that

462

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

Fig. 6. Immunofluorescence micrographs showing the co-localization of labelling for A1 R and CD26 in cell cultures of astrocytes. Overlapping immunoreactivity of CD26 and A1 R (A, B) and of CD26 and ADA (C, D) in cultures enriched with astrocytes. Bars: A: 20 ␮m, B: 10 ␮m.

of mammals (Ekström and Ebbesson, 1989) where an A1 R-mediated regulation of sleep and nociception has been shown (Tanase et al., 2001). The neuronal circuitry of the cerebellum is remarkably similar in all classes of vertebrates (Meek, 1992b) and in this area A1 Rs are known to modulate the release of glutamate from granule cells (Sanz et al., 1996; Schousboe et al., 1989). From the point of view of the evolution, the telencephalon of the teleosts is different from that of mammals, however, A1 Rs were found both in the goldfish telencephalon and in the telencephalic structures of the rat brain (Rivkees et al., 1995; Ciruela et al., 1995; Ochiishi et al., 1999). The A1 R appears to have been highly conserved during evolution, since immunoblot analysis using a monoclonal antibody for rat A1 R revealed an isoform of approximately 35 kDa which corresponds to that found in mammals (Ochiishi et al., 1999). Previous studies have shown that the A1 adenosine receptor appeared pharmacologically and biochemically similar to that found in mammals (Eschke et al., 2001; Cunha, 2001) and its neuromodulatory function in the

nervous system of lower vertebrate was comparable to that described in mammalian brain (Lucchi et al., 1992, 1994; Rosati et al., 1995). The presence of A1 R-immunoreactivity in neurons of cerebellum and optic tectum corresponds with the functional role of A1 adenosine receptors on the modulation of the Ca2+ -dependent neurotransmitter release observed in these goldfish brain areas (Poli et al., 1993, 2001). The immunopositivity for A1 Rs has been observed not only in neural component of the rhombencephalon and optic tectum slices, but also in the astroglial component, mainly in the radial glia (Kalman, 1998). The glial expression of A1 R has been confirmed using primary cultures of goldfish astrocytes. In the fish brain, radial glia are relatively abundant, and are considered phylogenetically ancient form of glia (Suarez et al., 1995). In the optic tectum, glial-radial extensions are closely associated with the dendrites of the neurons of the stratum periventricolaris, whose cell bodies show an intense staining for A1 Rs. Hence, a functional cross-talk between glia and adjacent neurons is well conceivable. A1 Rs localized in astrocytes from mammalian brain have been

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

implicated in neuroprotection against brain injures (Michel et al., 1999). The activation of A1 Rs stimulates astrocytes to produce trophic factors, such as nerve growth factor and S100␤ protein, which contribute to protect neurons against injury (Ciccarelli et al., 1999, 2001). These findings suggest that an analogous function, could be exerted by astrocytes in the brain of this teleost which is characterized by a high capacity to tolerate anoxia (Lutz et al., 1996). 4.2. Adenosine deaminase and adenosine deaminase-binding protein (CD26) Adenosine deaminase, an enzyme involved in purine metabolism which converts adenosine into inosine, is not only a cytosolic enzyme, but it is also an ecto-enzyme localized on the cell surface of many cells (Franco et al., 1986; Aran et al., 1991). It exhibits molecular characteristics and catalytic activity identical to the cytosolic ADA (Franco et al., 1997). Ecto-ADA is anchored to the cell surface by CD26 protein (Kameoka et al., 1993; Dong et al., 1996) and recently it has been shown that the ecto-ADA co-localizes with extracellular A1 Rs in cultured neurons from rat cerebral cortex (Ruiz et al., 2000) as well as in non-neural cells such as the smooth muscle cell line DDT1 MF-2 (Ciruela et al., 1996). Apart from degrading extracellular adenosine, the co-localization of A1 Rs and ADA forms the functional basis of the extra-enzymatic role of ecto-ADA in modulating ligand-induced signalling, desensitization and internalization of A1 Rs (Ciruela et al., 1996; Saura et al., 1998; Gines et al., 2001). Moreover, ecto-ADA is able to transmit signals when interacting with either A1 R or CD26 or both, thus acting as a co-stimulatory molecule in the regulation of CNS activity (Franco et al., 1997, 1998). In both membrane preparations and cytosol of goldfish brain the immunoblot analysis for ADA showed a single band (43 kDa) corresponding closely to that of mammals, whereas a solitary band (35 kDa) for A1 R was only shown in membranes. In the same A1 R-positive brain areas of goldfish, a marked expression of ADA was observed both in the neuronal and glial components. Moreover, the antibody anti-CD26 protein marked several populations of neurons in the rhombencephalon as well as the radial glia of optic tectum where immunostaining for ADA and A1 R was observed. We cannot assert that a co-localization between extracellular A1 R and ADA exists in the considered areas of the goldfish brain, since specific experiments were not performed. However, the following observations led us to suggest a probable functional interaction between the two proteins also in the neurons of goldfish as it has been established to exist in the neurons of mammals (Ruiz et al., 2000). The A1 receptor and adenosine deaminase were both expressed in goldfish brain membranes and a similar distribution of the immunoreactivity for A1 Rs and ADA was shown in the same regions of goldfish brain. In almost all brain areas expressing immunoreactivity for both A1 Rs and ADA, the anti-CD26 antibody showed a positive staining for CD26,

463

which is pivotal protein for anchoring ecto-ADA and is considered an important molecule that could allow ecto-ADA to act as a bridge between two cells expressing cell surface CD26 or between a cell expressing cell surface CD26 and another expressing A1 R (Franco et al., 1997). Indeed, the co-localization between A1 R, ADA and CD26 proteins is revealed in the glial component of the goldfish brain. The stratum marginalis of radial glia in the optic tectum presented immunoreactivity for A1 R, ADA and CD26, and in goldfish astrocyte cultures the immunohistochemical staining of A1 R, ADA and CD26, performed on the same cell population displayed a complete overlap of the distribution of the three antibodies.

5. Conclusion In conclusion, the immunohistochemical studies confirm our previous research on the distribution of A1 adenosine receptors, however, the parallel immunopositivity, at least in some discrete neuronal areas, for ADA and CD26 led us to hypothesize that a co-localization among A1 R, ecto-ADA and CD26 also exists in the neurons of goldfish since it has been established to exist in the neurons of mammals. Moreover, we have demonstrated for the first time, that A1 R, ecto-ADA and CD26 co-localization is present on the astroglial component of the goldfish brain. This raises the possibility that a similar situation is also shown in the glia of the mammalian brain.

Acknowledgements This work was supported by the Italian Ministry for University and Research (funds Ateneo 60% and Cofinanziamento 1999). We are grateful to Dr. T Shirao and Dr. H. Nakata for their technical suggestion about Western blotting and for the gift of the antibody directed against A1 receptor. References Aran, J.M., Colomer, D., Matutes, E., Vives-Corrons, J.L., Franco, R., 1991. Presence of adenosine deaminase on the surface of mononuclear blood cells: immunochemical localization using light and electron microscopy. J. Histochem. Cytochem. 39, 1001–1008. Ciccarelli, R., Di Iorio, P., Giuliani, P., D’Alimonte, I., Allerini, P., Caciagli, F., Rathbone, M.P., 1999. Rat cultured astrocytes release guanine-based purine in basal conditions and after hypoxia/ hypoglycemia. Glia 25, 93–95. Ciccarelli, R., Ballerini, P., Sabatino, G., Rathbone, M.P., D’Onofrio, M., Caciagli, F., Di Iorio, P., 2001. Involvement of astrocytes in purine-mediated reparative processes in the brain. Int. J. Dev. Neurosci. 19, 395–414 (Review). Ciruela, F., Casado, V., Mallol, J., Canela, E.I., Franco, R., 1995. Immunological identification of A1 adenosine receptors in brain cortex. J. Neurosci. Res. 42, 818–828. Ciruela, F., Saura, C., Canela, E.I., Mallol, J., Lluis, C., Franco, R., 1996. Adenosine deaminase affects ligand-induced signalling by interacting with cell surface adenosine receptors. FEBS 380, 219–223.

464

A. Beraudi et al. / Neurochemistry International 42 (2003) 455–464

Cunha, R.A., 2001. Adenosine as a neuromodulator and as a homeostatic regulator in the nervous system: different roles, different sources and different receptors. Neurochem. Int. 38, 107–125 (Review). Daddona, P.E., Kelley, W.N., 1977. Human adenosine deaminase: purification and subunit structure. J. Biol. Chem. 252, 110–115. Dong, R.P., Kameoka, J., Hegen, M., Tanaka, T., Xu, Y., Schlossman, S.F., Morimoto, C., 1996. Characterization of adenosine deaminase binding to human CD26 on T cells and its biologic role in immune responses. J. Immunol. 156, 1349–1355. Ekström, P., Ebbesson, S.O., 1989. Distribution of serotonin-immunoreactive neurons in the brain of sockeye salmon fry. J. Chem. Neuroanat. 2, 201–213. Eschke, D., Brand, A., Scheibler, P., Hess, S., Eger, K., Allgaier, C., Nieber, K., 2001. Effect of an adenosine A(1) receptor agonist and a novel pyrimidoindole on membrane properties and neurotransmitter release in rat cortical and hippocampal neurons. Neurochem. Int. 38, 391–398. Franco, R., Canela, E.I., Bozal, J., 1986. Heterogeneous localization of some purine enzymes in subcellular fractions of rat brain and cerebellum. Neurochem. Res. 11, 423–435. Franco, R., Casado, V., Ciruela, F., Saura, C., Mallol, J., Canela, E.I., Lluis, C., 1997. Cell surface adenosine deaminase: much more than an ecto-enzyme. Prog. Neurobiol. 52, 283–294 (Review). Franco, R., Mallol, J., Casado, V., Lluis, C., Canela, E.I., Saura, C., Blanco, J., Ciruela, F., 1998. Ecto-adenosine deaminase: an ecto-enzyme and a co-stimulatory protein acting on a variety of cell surface receptors. Drug Dev. Res. 45, 261–268. Gines, S., Ciruela, F., Burgueño, J., Casado, V., Canela, E.I., Mallol, J., Lluis, C., Franco, R., 2001. Involvement of caveolin in ligand-induced recruitment and internalization of A1 adenosine receptor and adenosine deaminase in an epithelial cell line. Mol. Pharmacol. 59, 1314–1323. Kalman, M., 1998. Astroglial architecture of the carp (Cyprinus carpio) brain as revealed by immunohistochemical staining against glial fibrillary acidic protein (GFAP). Anat. Embryol. (Berl.) 198, 409–433. Kameoka, J., Tanaka, T., Nojima, Y., Schlossman, S.F., Morimoto, C., 1993. Direct association of adenosine deaminase with a T cell activation antigen, CD26. Science 261, 466–469. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, L.J., 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265– 275. Lucchi, R., Poli, A., Traversa, U., Barnabei, O., 1992. Characterization of A1 adenosine receptors in membranes from whole goldfish brain. Comp. Biochem. Physiol. 102B, 331–336. Lucchi, R., Poli, A., Traversa, U., Barnabei, O., 1994. Functional adenosine A1 receptors in goldfish brain: regional distribution and inhibition of K+ -evoked glutamate release from cerebellar slices. Neuroscience 58, 237–243. Lutz, P.L., Nilsson, G.E., Perez-Pinzon, M.A., 1996. Anoxia tolerant animals from a neurobiological perspective. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 113, 3–13 (Review). Martin, M., Huguet, J., Centelles, J.J., Franco, R., 1995. Expression of ecto-adenosine deaminase and CD26 in human T cells triggered by the TRC-CD3 complex: possible role of adenosine deaminase as co-stimulatory molecule. J. Immunol. 155, 4630–4643. Meek, J., 1992b. Comparative aspects of cerebellar organization from mormyrids to mammals. Eur. J. Morphol. 30, 37–51 (Review). Michel, P.P., Marien, M., Ruberg, M., Colpaert, F., Agid, Y., 1999. Adenosine prevents the death of mesencephalic dopaminergic neurons by a mechanism that involves astrocytes. J. Neurochem. 72, 2074–2082. Nagy, J.I., Yamamoto, T., Uemura, H., Schrader, W.P., 1996. Adenosine deaminase in rodent median eminence: detection by antibody to the mouse enzyme and co-localization with adenosine deaminase-complexing protein (CD26). Neuroscience 73, 459–471.

Nakata, H., 1993. Development of an antiserum to rat brain A1 adenosine receptor: application for immunological and structural comparison of A1 adenosine receptors from various tissues and species. Biochim. Biophys. Acta 1177, 93–98. 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 receptor-specific monoclonal antibody. J. Comp. Neurol. 411, 301–316. Poli, A., Lucchi, R., Zottini, M., Traversa, U., 1993. Adenosine A1 receptor-mediated inhibition of evoked glutamate release is coupled to calcium influx decrease in goldfish brain synaptosomes. Brain Res. 620, 245–250. Poli, A., Di Iorio, P., Beraudi, A., Notari, S., Zaccanti, F., Villani, L., Traversa, U., 2001. The calcium-dependent [3 H] acetylcholine release from synaptosomes of brown trout (Salmo trutta) optic tectum is inhibited by adenosine A1 receptors: effects of enucleation on A1 receptor density and cholinergic markers. Brain Res. 892, 78– 85. Rivkees, S.A., Price, S.L., Zhou, F.C., 1995. Immunohistochemical detection of A1 adenosine receptors in rat brain with emphasis on localization in the hippocampal formation, cerebral cortex, cerebellum, and basal ganglia. Brain Res. 677, 193–203. Rosati, A.M., Traversa, U., Lucchi, R., Poli, A., 1995. Biochemical and pharmacological evidence for the presence of A1 but not A2 adenosine receptors in the brain of the low vertebrate teleost Carassius auratus (goldfish). Neurochem. Int. 26, 411–423. Ruiz, M., Escriche, M., Lluis, C., Franco, R., Martin, M., Andrei, A., Ros, M., 2000. Adenosine A1 receptors in cultured neurons from rat cerebral cortex: co-localization with adenosine deaminase. J. Neurochem. 75, 656–664. Sanz, J.M., Vendite, D., Fernandez, M., Andres, A., Ros, M., 1996. Adenosine A1 receptors in cultured cerebellar granule cells: role of endogenous adenosine. J. Neurochem. 67, 1469–1477. Saura, C., Mallol, J., Canela, E., Lluis, C., Franco, R., 1998. Adenosine deaminase and A1 adenosine receptors internalize together following agonist-induced receptor desensitization. J. Biol. Chem. 273, 17610– 17617. Schousboe, A., Frandsen, A., Drejer, J., 1989. Evidence for evoked release of adenosine and glutamate from cultured cerebellar granule cells. Neurochem. Res. 14, 871–875. Schrader, W.P., West, C.A., Strominger, N., 1987. Localization of adenosine deaminase and adenosine deaminase complexing protein in rabbit brain. J. Histochem. Cytochem. 35, 443–451. Siebenaller, J.F., Murray, T.F., 1986. Phylogenetic distribution of [3 H] cyclohexyladenosine binding sites in nervous tissue. Biochem. Biophys Res. Commun. 137, 182–189. Sivron Jeserich, G., Nona, S., Schwartz, M., 1992. Characteristics of fish glial cells in culture: possible implications as to their lineage. Glia 6, 52–66. Suarez, I., Bodega, G., Rubio, M., Fernandez, B., 1995. Evolution of astrocytes in vertebrate CNS. In: Vernadakis, A., Roots, B. (Eds.), Neuronglia Interrelations During Phylogeny. Neuroglia, Humana press, Totowa, NJ, pp. 59–77. Tanase, D., Martin, W.A., Baghdoyan, H.A., Lydic, R., 2001. G protein activation in rat ponto-mesencephalic nuclei is enhanced by combined treatment with a mu opioid and an adenosine A1 receptor agonist. Sleep 24, 52–62. Yamamoto, T., Geiger, J.D., Daddona, P.E., Nagy, J.I., 1987. Subcellular, regional and immunohistochemical localization of adenosine deaminase in various species. Brain Res. Bull. 19, 473–484.