BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL

Transcription

1 Vol. 44, No. 6, May 1998 BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL Pages CIIARACTERISATION OF L-[3H]GLUTAMATE BINDING TO FRESH AND FROZEN CRUDE PLASMA MEMBRANES ISOLATED FROM CEREBRAL CORTEX OF ADULT RATS. Emanuelli, T."'b; Antunes, V.F. a and Souza, D.O.G. 1,a "Departamento de Bioquimica, Instituto de Ci~ncias Bhsicas da Saflde, Universidade Federal do Rio Grande do Sul, , Porto Alegre, RS, Brasil. bdepartamento de Tecnologia e Ci6ncia dos Alimentos, Centro de Ci~ncias Rurais, Universidade Federal de Santa Maria, , Santa Maria, RS, Brasil. Received December 15, 1997 Received after revision, February 3, 1998 Summary Specific [3H]glutamate binding to fresh crude plasma membranes (CPMs) was compared with binding to frozen CPMs and the optimal conditions for the binding to frozen CPMs isolated from cerebral cortex of adult rats were determined. Freezing reduced [3H]glutamate binding (3.5-fold), and pre-incubation of previously frozen membranes followed by three washes increased binding (4.5-fold) when compared to fresh samples. CPMs washed once, pre-incubated at 37~ and washed 3 times was adopted as the most adequate condition for the binding assay of frozen membranes. In a Cl--containing medium, [H]glutamate binding (Bmax=97.9 pmol/mg, Kd= nm) to this frozen CPM preparation was significantly displaced by excess quisqualic acid (QA) (65%), L-2-amino-4-phosphonobutyric acid (L-AP4) (35%), trans-laminocyclopentane-l,3-dicarboxylate (1S,3R-ACPD) (25%) and alfa-amino-3-hydroxy-5-methyl- 4-isoxazolepropionate (AMPA) (25%). In a Cl--free medium, binding (Bma = pmol/mg, 311 nm) was significantly displaced by QA (45%), L-AP4 (25%), ACPD (25%), AMPA (25%), kainic acid (2%) and N-methyl-D-aspartate (15%). Key Words: chloride-dependent [3H]glutamate binding, chloride-independent [3H]glutamate binding, fresh crude plasma membranes, frozen crude plasma membranes, rat cerebral cortex, freezing. Introduction Glutamate is the major excitatory neurotransmitter in the mammalian central nervous system (1). Glutamate receptors have been classified into two major classes: ionotropic (ligand-gated ion channels) and metabotropic (G protein-coupled receptors) receptors. Ionotropic glutamate receptors can be distinguished pharmacologically by their specific binding of the Abbreviations: CPM, crude plasma membrane, NMDA, N-methyl-D-aspartate; KA, kainic acid; AMPA, affaamino-3-hydroxy-5-methyl-4-isoxazolepropionate; 1S,3R-ACPD, trans-l-aminocyclopentano-l,3-dicarboxilate; L- AP-4, L-2-amino-4-phosphonobutyric acid, QA, quisqualic acid. Corresponding author. Tel. (55) (51) ; FAX. (55) (51) /98/ / Copyright by Academic Press Australia All rights of reproduction in any fi~rm reserved.

2 BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL agonists N-methyl-D-aspartate (NMDA), kainic acid (KA) and alfa-amino-3-hydroxy-5-methyl-4- i soxazolepropionate (AMPA) (1). Metabotropic receptors are activated by ligands such as trans-laminocyclopentane-l,3-dicarboxylate (1S,3R-ACPD), L-2-amino-4-phosphonobutyric acid (L- AP4), ibotenate and quisqualic acid (QA) (1). An understanding of the glutamate receptor system is crucial to an understanding of basic brain functions, such as learning and memory (2); for the rational treatment of acute pathological insults and neurodegenerative disorders associated with glutamatergic dysfunction (3) and in the searh for new therapeutic drugs (4). Binding studies are a standard procedure for the in vitro analysis of the interactions of neurotransmitters with their receptors. Hence, there has been extensive investigation of [3H]glutamate binding to cerebral plasma membranes (5). Even though [3H]glutamate binding to frozen membranes seems to be lower than to fresh membranes (6, 7), there has been some controversy as to the most adequate procedure for plasma membrane preparation for [3H]glutamate binding. Some authors perform [3H]glutamate binding to fresh membranes (8, 9), while others perform [3H]glutamate binding to membranes frozen during or after the membrane preparation (1, 11). Moreover, there are few studies comparing [3H]glutamate binding to fresh and frozen synaptic plasma membranes (6, 7, 12) or investigations of the optimal conditions for glutamate binding to frozen synaptic plasma membranes. Freezing procedures are useful since they allow the storage of samples for later assay, and reduce the time taken by membrane preparation on the day of the binding assay This investigation involved the following: (i) comparison of [3H]glutamate binding to fresh crude plasma membranes (CPMs) to [3H]glutamate binding to frozen CPMs, (ii) establishment of the most adequate procedure for preparing frozen CPMs for [3H]glutamate binding, (iii) determination of optimal conditions for a [3H]glutamate binding assay and (iv) characterisation of this binding to frozen CPMs isolated from rat cerebral cortex. Material and Methods Materials: L-[3H]Glutamate (48 Ci/mmol) was purchased from Amersham International, U.K. AMPA, NMDA and QA were from Research Biochemicals International (Natick, MA, USA). Glutamic acid was purchased from Merck (Darmstadt, Germany). KA was obtained from Sigma (St. Louis, MO, USA). L-AP4 was obtained from Tocris Cookson Ltd (Bristol, UK). All other chemicals were of analytical reagent grade. Animals: Male wistar rats (6-9-day-old) from our own breeding colony were used. They were maintained at 25~ on a 12-h light/12-h dark cycle. Food and water were given ad libitum. 1266

3 BIOCHEMISTRYGnd MOLECULAR BIOLOGY INTERNATIONAL Membrane Preparation_; Preparation of fresh cortical crude plasma membranes (fresh CPMs): Crude plasma membranes (CPMs) were prepared using the procedure of (13), with minor modifications, as follows. Rats 9 were killed by decapitation, cerebral cortices were removed and homogenised in 2 volumes (ml per gram of wet tissue) of.32 M sucrose, 1 ~ MgCI2 and 1 mm Tris-HC1 buffer, ph 7.4, using a hand-operated glass homogeniser. The homogenate was centrifuged at 1, g for 15 rain, and the pellet was resuspended in 2 volumes (ml per gram of original wet tissue) of the same buffer and centrifuged again. The second pellet was discarded and the supernatant fractions were pooled and centrifuged at 27, g for 15 rain. The resulting pellet was lysed in 2 volumes of l mm Tris-HCl buffer, ph 7.4 for 3 rain, and centrifuged at 27, g for 15 min. This pellet was washed three times in 2 volumes of 1 mm Tris-HCl, ph 7.4, at 27, g for 15 min. The final pellet was resuspended in 1 mm Tris-HC1, ph 7.4 in order to yield a final protein concentration of 1-2 mg/rnl and was used for the binding assays. All centrifugation steps were carried out at 4~ in a Sorvall RC 5B centrifuge, with an SS-34 rotor. Preparation of frozen cortical crude plasma membranes (frozen CPMs): Fresh CPMs prepared as described above were frozen at - 7~ for at least 24 h and were not stored for more than two months. On the day of binding assay, the membranes were rapidly thawed in a water bath (37~ homogenised with 3 volumes of 1 mm Tris-HCl, ph 7.4, and centrifuged at 27, g for 15 min. The resulting pellet was resuspended in 3 volumes (ml per ml of thawed membrane) of the same buffer, pre-incubated at 37~ for 3 min, and centrifuged at 27, g for 15 rain. The pellet was resuspended in 3 volumes of 1 mlvl Tris-HC1 or Tris-Acetate buffer and washed 1 to 6 times in 3 volumes of the same buffer, and centrifuged at 27, g for 15 min. The final pellet was resuspended in the same buffer in order to yield a protein concentration of 1-2 mg/ml and was used for the binding assays. All centrifugation steps were carried out at 4~ in a Sorvall RC 5B centrifuge. [3H]Glutamate binding assay: Unless otherwise indicated, a fixed concentration of 4 nm [3H]glutamate was used and incubations were started by the addition of 1-2 ~g of membrane protein to a medium containing 5 mm Tris/HCl or Tris/Acetate buffer, ph 7.4. Following 15-3 minutes of incubation at 3~ bound and free ligands were separated by centrifugation at 15,8 g for 15 min. Preliminary experiments in which centrifugation time was assayed had shown that all glutamate binding membranes sedimented under these conditions (data not shown). The supernatant fractions were aspirated and the pellets were quickly and carefully rinsed with ice cooled distilled water, and solubilised with.3 ml.1% sodium dodecyl sulfate (w/v) overnight and radioactivity was measured. Specific glutamate binding was calculated as the difference between total binding and unespecific binding which was measured in the presence of a 1,-fold excess of unlabelled L-glutamate. Unespecific binding typically amounted to 1-2% of the total binding. Protein Assay: Protein concentration was measured according to the method of (14) with bovine serum albumine as the standard. Results and Discussion Comparison of glutamate binding to fresh and frozen CPMs and establishment of the most adequate procedure for preparing frozen CPMs for [3H]glutamate binding: In accord with published data the results reported here (Figure 1) demonstrated that [SH]glutamate binding to 1267

4 BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL.~ 3 E o 24 E cl. o a\b cd e f g h i J~ frozen membranes FIGURE 1 - Glutamate binding to fresh and frozen CPMs. Binding of [3H]glutamate (4 nm) to CPMs prepared from cerebral cortex of adult rats was assayed in Tris-HCl buffer, at 3~ for 15 rain. Reaction was terminated by centrifugation at 15,8 g for 15 min. (a) fresh membrane, (b) frozen nonwashed membrane, (c) frozen membrane washed once, (d) frozen membrane washed once, pre-incubated at 37~ for 3 min, centrifuged at 27, g for 15 min and resuspended in 1 mm Tris-HC1, ph 7.4 (e) frozen membrane, washed once, pre-incubated and washed once more, (f) frozen membrane, washed once, pre-incubated and washed twice, (g) frozen membrane, washed once, pre-incubated and washed three times, (h) frozen membrane, washed once, preincubated and washed four times, (i) frozen membrane, washed once, pre-incubated and washed five times, (j) frozen membrane, washed once, pre-incubated and washed six times, (k) frozen membrane washed once, pre-incubated at 37C for 3 min in the presence of 5% triton X-1 and washed three-fold. Results are means+s.e, (n=6). * Significantly different from fresh membrane (a) (p<.5). " Significantly different from frozen membrane, washed once, pre-incubated and washed three times (g) (p<.5). frozen and nonwashed membranes (in a Cl--containing buffer) was drastically reduced in comparison with binding data for fresh membrane preparations (6, 7). The decrease may be attributed to the break up of membrane vesicles and the release of endogenous components such as glutamate. Washing once the previously frozen CPMs was found to increase (17%) [3H]glutamate binding over the binding of glutamate to fresh CPMs, Incubation of frozen CPMs for 3 min at 37~ followed by up to three washes was found to further increase glutamate binding (6%) (Figure 1). Additional washings did not promote any further significant increase of glutamate binding to frozen CPMs. The literature values are unsound in that [3H]glutamate binding has been reported using either fresh, frozen and nonwashed, or frozen and washed CPMs (9, 11, 15). However, comparative studies on [3H]glutamate binding have always compared binding data using fresh membranes with that of frozen and nonwashed membranes (6, 7, 12). No previous study of [3H]glutamate binding to frozen membranes, has reported an increase in glutamate binding in comparison with that of fresh membranes (1, 11, 12). In this study a 1268

5 BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL comparison of glutamate binding to fresh or frozen and thoroughly washed membranes was made and a significantly higher [3H]glutamate binding to pre-incubated and thoroughly washed frozen membranes when compared to fresh or frozen nonwashed CPMs was noted. Enhanced [3H]glutamate binding is likely to be due to the removal of endogenous components (released during the freeze-thawing procedure) such as glutamate, soluble enzymes and/or other unknwon interfering compounds. The pre-incubation of the membranes with a detergent such as Triton X- 1, followed by a regime of washing has been employed in binding assays (16) to deplete the membranes of endogenous ligands and increase specific binding. Accordingly, the effects of Triton X-1 (.5%) on the binding of [3H]glutamate to frozen CPMs was examined. Although.5% Triton X-1 promoted an increase of 25% in [3H]glutamate binding when compared to frozen CPMs pre-incubated in the absence of Triton X-1 and washed 3 times (Figure 1), it also markedly increased the toss of protein during the washing procedures (65-75% vs. 2-5% loss). Moreover, it was reasoned that detergent treatment may disturb the native conformation of the plasma membranes. Hence in this study the following conditions were adopted for [3H]glutamate binding to frozen membranes, CPMs were washed once, pre-incubated in the absence of Triton X- 1 at 37~ for 3 min, centrifuged at 27, g for 15 min and washed three times. This procedure did not result in any reduction in the time of membrane preparation on the day of the binding assay, however, it yielded a [3H]glutamate binding level significantly higher than that obtained using fresh membranes. Determination of optimal conditions for glutamate binding: [3H]glutamate binding (4 nm) to frozen CPMs was linearly related to the amount of protein in the assay system in the range O. 1 to.8 mg protein/ml and reached equilibrium within 3 min of incubation (data not shown). Kinetic characterisation of [3H]glutamate binding to frozen CPMs: Specific binding was saturable over the range 2-3, nm (Figure 2). Scatchard plot analysis of the data obtained in the presence of CI- ions revealed a homogenous population of binding sites (Bmax pmol/mg protein and Kd nm, mean+s.e., n=4) (Figure 2). The Kd value is in agreement with a previous report of [3H]glutamate binding to frozen rat cortical membranes in a medium with a similar ionic composition (17). However, the Bmax (98 pmol/mg) was much higher than that reported by (17) (12 pmol/mg) for rat cortical synaptic membranes using a Cl--containing buffer. This discrepancy may be due to differences in the particulate fractions used or in the preparation procedure. Misinterpretation of binding data may result from the binding of the ligand 1269

6 BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL 12 E 1 o s --~ 6 "~ 4._ X 2 m l.35! f_-? _o28 2"I I ~".14 /i ~ -~ ~,~_~ - - i , ! Bound (pmol/mg) glutamate (nm) FIGURE 2 - Saturation curves of [3H]glutamate specific binding to frozen CPMs. Membranes were thawed, washed once, pre-incubated at 37~ for 3 min, washed three times and used for [3H]glutamate binding. Binding was performed in Tris-HC1 (e) or Tris-Acetate (+), at 3~ for 3 min. The reaction was terminated by centrifugation at 15,8 g for 15 rain. Saturation curves were obtained using [3H]glutamate in a concentration range from 2 nm to 3, nm. The inset shows a scatchard analysis of equilibrium binding data. Results are means+se. (n=4). to transport sites. In this study the possibility of glutamate binding to its transport sites was ruled out since binding studies were also performed in the absence of Na* and CI- ions (Tris-Acetate buffer), which are known to be essential for glutamate binding at the transport site (18, 19, 2). In the absence of C1- ions (Tris-Acetate) the Kd was not significantly modified (311.+_52.14 nm) (mean+s.e., n=4). In contrast, Bmax was decreased to pmol/mg protein (mean+s.e., n=4). These results suggest the presence of a C1--dependent population of [3H]glutamate binding sites in the frozen CPMs. This population of sites may be related to the Cl--dependent glutamate carrier (19). Although C1--dependent glutamate binding has earlier been claimed to be eliminated by freezing (7, 12), experimental evidence has indicated the presence of a C1--dependent population of [3H]glutamate binding sites in previuosly frozen cerebral membranes (1). The dissociation constant of glutamate binding in the absence of CI- ions was similar to the value obtained in the presence of CI- ions and to the one reported in (21) for rat cortical membranes. These results indicated a similarity in the affinity of [3H]glutamate for both Cl--dependent and independent population sites. Pharmacological characterisation of [3H]glutamate binding to frozen CPMs: The pharmacological profile of [3H]glutamate binding assayed in the presence of C1- ions was slightly different from the one observed in the absence of these ions. In a C1--containing buffer, AMP& QA, ACPD and L- AP4 receptor sites appeared to contribute to [3H]glutamate binding (Figure 3A). While, in a C1-- free buffer the contribution of QA for [3H]glutamate binding was significantly reduced and NMDA 127

7 BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL o :~ ~ 6 < z <: < -5 A I B FIGURE 3 - Displacement of specific [3H]glutamate binding by various glutamatergic agonists. Binding of 4 nm [3H]glutamate was performed using frozen CPMs prepared from cerebral cortex of adult rats. Membranes were thawed, washed once, pre-incubated at 37~ for 3 min, washed three-fold and used for glutamate binding. Displacing compounds (16 gm) were pre-incubated with frozen CPMs in Tris-HC1 (A) or Tris-Acetate (B) for 5 rain at 3~ The incubation was initiated by the addition of [3H]glutamate and continued for 3 min at 3~ Reaction was terminated by centrifugation at 15,8 g for 15 min. Results are means+s.e (n=3). Two-way ANOVA (2 buffer x 7 displacing compound) revealed a significant main effect of ligand (F(6,35) = 33.61, p<.1) and a significant buffer x displacing compound interaction (F(6,35)= 2.49, p<.5). * Significantly different from control (p<.5). and KA receptor contributions appeared (Figure 3B). This is consistent with a reduction of [3H]glutamate binding to Cl--dependent carrier site which is known to be sensitive to QA (19). The results of this study are confirmed by data in the literature and demonstrate that freezing decreased [3H]glutamate binding. Furthermore, for the first time it is shown that freezing enhanced [3H]glutamate binding when the membranes were pre-incubated and washed after thawing. Additionally, optimal conditions for the measurement of [3H]glutamate binding to previously frozen rat cortical CPMs are described. Acknowledgements We thank Andrea Regner and Dr. JoAo Batista Teixeira da Rocha for a critical appreciation of the manuscript. This work was supported by CNPq (PRONEX grant /96 to D.O.G Souza). Vicente Freitas Antunes is the recipient of CNPq Scientific Initiation Fellowships (Programa Institucional UFRGS-1996 and 1997). References I. Hollmann, M. and Heinemann, S. (1994) Annu. Rev. Neurosci. 17, McEntee, W.J. and Crook, T.H. (1993)Psychopharmacology. 11, Lipton, S.A. and Rosenberg, P.A. (1994) New Eng. J. Med. 33,

8 BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL 4. Elisabetsky, E.; Marschner, J. and Souza, D.O. (1995)Neurochem. Res. 2, Collingridge, G.L. and Lester, R.A. (1989) Pharmacol. Rev. 4, Foster, A.C. and Fagg, G.E. (1984) Brain Res. Reviews. 7, Wu, K.; Carlin, R. and Siekevitz, P. (1986) J. Neurochem. 46, Rao, V.L.R. and Murthy, Ch.R.K. (1993) Biochem. Mol. Biol. Int. 3, Rubin, M.A.; Medeiros, A.C.; Rocha, P.C.B.; Livi, C.B.; Ramirez, G.; Souza, D.O. (1997) Neurochem Res. 22, Hollmann, M. and Seifert, W. (1989) J. Neurochem. 53, Schroder, H.; Becker, A. and Lossner, B. (1993) J. Neurochem 6, Fagg, G.E.; Mena, E.E.; Monaghan, D.T. and Cotman, C.W. (1983) Neurosci. Lett. 38, t3. Paz, M.M.; Ramos, M.; Ramirez, G. and Souza, D. (1994) FEBS Lett. 355, Lowry, O.H., Rosebrough, N.J.; Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem.. 193, Cowburn, R.F.; Hardy, J.A. and Roberts, P.J. (1988) J. Neuroehem. 5, Wong, D.T. and Horng, J.S. (1979) J. Neurochem. 32, Iwata, H.; Yamagami, S. and Baba, A. (1982) J. Neuroehem. 38, Kessler, M.; Baudry, M. and Lynch, G. (1987) Neurosci. Lett. 81, Anderson, K.J.; Monaghan, D.T.; Bridges, R.J.; Tavoularis, A.L. and Cotman, C.W. (199) Neurosci., 38, Worral, D.M. and Williams, D.C. (1994) Biochem. J. 297, Sanderson, C. and Murphy, S. (1982) Develop. Brain Res., 2, I272