- Email: [email protected]
Localization of the Cl−-ATPase activity on NAP-22 enriched membrane microdomain (raft) of rat brain
Neuroscience Letters 362 (2004) 158–161 www.elsevier.com/locate/neulet
Localization of the Cl2-ATPase activity on NAP-22 enriched membrane microdomain (raft) of rat brain Shohei Maekawa*, Katsutoshi Taguchi Division of Bioinformation, Department of Biosystems Science, Graduate School of Science and Technology, Kobe University, Kobe 657-8501, Japan Received 25 February 2004; received in revised form 11 March 2004; accepted 12 March 2004
Abstract Much attention has been paid to the membrane microdomain enriched in cholesterol and sphingolipids called raft. In the central nervous system, however, the physiological role of this domain is not so evident at present, partly because of the complexity of the protein components in the raft fraction. In this study we surveyed ATPase activities in the raft fraction obtained from the synaptic plasma membrane of rat brain and found the enrichment of an ethacrynic acid-sensitive ATPase (Cl2-pump) activity. Immunoprecipitation experiments using antibodies to raft-localized proteins showed the co-precipitation of the ATPase activity with NAP-22, a major raft-localized protein. This result suggests the participation of the raft in the regulation of ion transport in addition to the presence of heterogeneity of raft domains in neurons. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Cl2-ATPase; Raft; Neuron; NAP-22; Cholesterol
Lateral heterogeneities in the classical fluid-mosaic model of cell membranes are now envisaged as caveloae-like microdomains or ‘rafts’, that are enriched in cholesterol, glycosphingolipids, and specific membrane proteins including glycosylphosphatidylinositol (GPI)-anchored proteins [1,3,9,18]. Since many signal-transducing molecules such as trimeric G proteins, protein tyrosine kinases, cytoskeletal proteins, and calmodulin binding proteins are present in this region, raft is considered as a signal-transducing region of the cell [1,9 – 13]. From the preferential transport of GPIanchored proteins to the apical region of the epithelial cells, the raft is also considered to be important to the establishment and maintenance of cell polarity [18]. Elucidation of the molecular interaction in this region is, hence, important not only to understand the molecular mechanisms of signal transduction in cells but also to understand the intracellular sorting mechanisms of this region. Raft is generally described as a Triton-insoluble fraction of low density, i.e. a fraction recovered in a low buoyant density fraction after Triton treatment and sucrose density gradient centrifugation. During this procedure, Triton-insoluble lipid domains are thought to assemble to make vesicles, and proteins having affinities to these lipids are recovered in this * Corresponding author. Tel./fax: þ81-78-803-6507. E-mail address: [email protected] (S. Maekawa).
fraction. In the brain-derived raft, specific localization of GPI-anchored proteins, Src family protein kinases, trimeric G proteins, and neuron-specific calmodulin binding proteins (NAP-22 and GAP-43) was reported [9,10,13]. Among these proteins, NAP-22 is very important in considering the organization of the neuronal raft domains, because this protein is prominent in amount and is neuron-specific, has a cholesterol binding activity, and induces a cholesterol enriched domain in vitro and in vivo [9,13]. In the cell membrane many energy-dependent transporters perform their functions to sustain the cellular function. One of the main transporters is the Naþ, Kþ-ATPase, which maintains the gradients of Naþ and Kþ ions between the cell membrane and is sensitive to ouabain. The plasma membrane Ca2þ-pump is another ATPase which excludes Ca2þ ions and the activity is enhanced by Ca2þ-calmodulin [14]. In addition to these proteins, Inagaki and her colleagues identified a Cl2-pump as an ethacrynic acid (EAA)-sensitive ATPase [5,7,17]. The Cl2-pump and a brain type Kþ/Cl2 cotransporter are known to develop an inwardly directed Cl2 gradient which enables hyperpolarizing inhibitory postsynaptic potentials generated by transmitter-operated Cl2 channel opening [5,7,17]. In order to know the physiological contribution of raft on the cell function, the ATPase activities in the raft fraction
0304-3940/03/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.03.020
S. Maekawa, K. Taguchi / Neuroscience Letters 362 (2004) 158–161
were surveyed. The presence of the ouabain-sensitive ATPase and an enrichment of EAA-sensitive ATPase activity were observed. The preparation of the synaptic plasma membrane fraction (SPM) from 6-week-old rat (Wistar) brains and the detergent-insoluble low-density membrane fraction (raft) was prepared from SPM using 1% Triton X-100 (Tx-LDM) or 0.5% Brij 96 (Br-LDM) as described previously [8,19]. In order to measure the activity distribution during tissue fractionation, rat forebrain (3week-old) was homogenized in 8 vol of 0.25 M sucrose, 10 mM Tris –HCl, 1 mM EDTA, and 1 mM phenylmethylsulfonyl fluoride (PMSF) (pH 7.4). After centrifugation at 1000 £ g for 10 min, the supernatant was saved. The pellet fraction (P1) was homogenized again in 4 vol of the above solution and re-centrifuged. The recovered supernatants were mixed and centrifuged at 10,000 £ g for 20 min. The pellet fraction (P2) was saved and the supernatant was further centrifuged at 100,000 £ g for 30 min to recover the pellet fraction (P3). The sensitivity of the ATPase activity to EAA was most prominent in the P3 fraction as described previously [5]. The low-density detergent-insoluble membrane fraction (raft), the high-density detergent-insoluble fraction (HDM), and the detergent soluble fraction (BS) were hence prepared from the P3 fraction using 0.5% Brij 96 as described previously [12]. The raft fraction was prepared from the P3 fraction and further extracted with 0.5% MEGA10 (w/v) in 10 mM Tris– HCl, 0.14 M NaCl, 1 mM MgCl2, and 0.2 mM EGTA (pH 7.4) at 37 8C for 20 min. After centrifugation at 100,000 £ g for 30 min, the supernatant was recovered. The supernatant was incubated for 2 h at 4 8C with CNBr-activated Sepharose CL4B beads on which several monoclonal antibodies were coupled. After washing the beads three times with a solution described above (without MEGA10), the beads were washed once with a medium for the ATPase assay (described below) without ATP. The ATPase activity was then measured with the addition of ATP or the bound proteins were eluted through incubation with 1 vol of 2 £ SDS-sample buffer (without mercaptoethanol) at 56 8C for 20 min. The Naþ, Kþ-ATPase was assayed in 10 mM Tris–HCl (pH 7.4), 5 mM MgCl2, 1 mM EDTA, 0.1 M NaCl, 10 mM KCl, 0.5 mM ATP, and 0.02 mg/ml protein, with or without 1 mM ouabain at 37 8C. Naþ, Kþ-ATPase activity was calculated by subtracting the activity in the presence of 1 mM ouabain from the total ATPase activity. The difference between the ouabain-insensitive activities in the presence or absence of 0.3 mM EAA was designated as the Cl2-ATPase activity [5,17]. The liberated phosphate was measured to assay the ATPase activity with the malachite green method [16]. SDS-PAGE analysis, Western blotting, and protein determination were performed as described previously [10 – 12]. Production of monoclonal antibodies against NAP-22, GAP-43, and F3 was described previously [9,10, 12]. At first, we assayed the hydrolysis activities of Triton-
159
raft and Brij-raft using ATP, GTP, ADP, and AMP as the substrates. Both rafts showed high activities to hydrolyze ATP and AMP. Since brain contains the GPI-anchored form of 50 -nucleotidase, the AMPase activity observed here could be ascribed to this enzyme [1,4,13]. In order to address the molecular species of raft-localized ATPases, the effects of various specific inhibitors were studied. Bafilomycin A1 (10 mM, V-ATPase), 2,3-butanedione-2-monoxime (10 mM, myosin ATPase), EGTA (2 mM, Ca2þ-ATPase), and 4hydroxynonenal (50 mM, ecto-ATPase) showed little effects on the activity (data not shown) [4,14 – 16,20]. On the other hand, ethacrynic acid (0.3 mM) and ouabain (1 mM) inhibited the activity to some extent. Fig. 1 shows the effect of EAA on SPM and SPM-derived rafts. The ATPase activity of Brij-raft was reduced to about 60% in the presence of EAA. The inhibitory effect of EAA was also observed in Triton-raft, although the effect was not so high. In contrast, EAA showed little effect on the ATPase activity of SPM. This result shows the enrichment of the EAA-sensitive (Cl2-pump) activity in the raft fraction. The ouabain-sensitive ATPase activity, in contrast, showed no specific localization on raft and further Western blotting using an antibody to this enzyme confirmed this result (data not shown). In order to analyze the localization of the EAA-sensitive ATPase in raft, the P3 fraction was further fractionated using Brij 96 to obtain Brij-soluble fraction (BrS), raft (Br-LDM), and Brij-insoluble high-density fraction (Br-HDM), and the ATPase activities were measured. The ouabain-sensitive ATPase activity was observed in both the BrS and the B-LDM fraction (data not shown). The EAA-sensitive ATPase was, in contrast, largely recovered in the Br-LDM (raft) fraction. The Br-HDM showed a substantial ATPase activity. This activity, however, showed poor EAA sensitivity (Fig. 2). Inagaki and her colleagues showed the solubilization of Cl2-pump with a detergent, MEGA10 [5,7]. Since the solubilized activity was recovered in a high molecular weight range in a native gel
Fig. 1. Presence of an EAA-sensitive ATPase in the raft (LDM) fraction obtained from SPM fraction using Triton X-100 or Brij 96. The ATPase activity was assayed in 10 mM Tris –HCl, 5 mM MgCl2, 1 mM EDTA, 0.1 M NaCl, 10 mM KCl, 0.5 mM ATP, 1 mM ouabain, and 0.02 mg/ml protein at 37 8C for 10 min in the presence or absence of 0.3 mM EAA. Bars represent SD (n ¼ 4). *P , 0:05.
160
S. Maekawa, K. Taguchi / Neuroscience Letters 362 (2004) 158–161
electrophoresis, they suggested an oligomeric nature of the enzyme [7]. The result described here suggests the association of the enzyme in raft in a multimolecular complex. The molecular characterization of brain-derived raft showed the localization of F3, NAP-22, GAP-43, and trimeric G protein Go as the major raft components [9,10]. Since the presence of a variety of rafts with different protein and lipid components is known, further studies on the colocalization the Cl2-pump with these proteins is of great importance. Immunoprecipitation experiments using monoclonal antibodies to these major proteins were, hence, applied after the partial solubilization of the raft with MEGA10. The SDS-PAGE analysis of the immunoprecipitated samples showed the specific recovery of the antigens but not other major proteins (Fig. 3B). This result shows that the raft proteins solubilized with MEGA10 are separable from each other. The ATPase analysis of these fractions showed a recovery of the EAA-sensitive ATPase activity in the NAP-22 immunoprecipitate but not in the other immunoprecipitates (Fig. 3A). Since raft is a lipid-based structure and the detergent only partially breaks down the raft through partial extraction of lipids, the recovery of the ATPase in a NAP-22 enriched raft does not directly mean an interaction between NAP-22 and the ATPase. Co-precipitation, however, suggests the possibility that these proteins localize in the same raft within the neuron. Since the molecular cloning of the Cl2-pump has not yet been done, the precise comparison of the localization pattern of the Cl2-pump with that of NAP-22 is impossible at present [6]. A rapid increase of the Cl2-ATPase activity during the neonatal stages is observed and this pattern is very similar to that of NAP-22 [7,9,17]. The localization of phosphatidylinositol 40 ,50 -bisphosphate (PIP2) in raft is well recognized and recent studies showed the localization of a novel type of phosphatidylinositol 4-kinase (PI-4K II) in the raft fraction [1,2,18]. The raft is therefore not only the storehouse of
Fig. 3. Immunoprecipitation of MEGA10 solubilized raft fraction using monoclonal antibodies to NAP-22 (NAP), GAP-43 (GAP), F3 (F3), and Go a subunit (Go). The antibody coupled gels were mixed with the fraction at 4 8C for 2 h. After washing the ATPase activity of the gels was assayed (A) in the presence or absence of EAA. Bars represent SD (n ¼ 4). *P , 0:05. The bound proteins were eluted from the gels with incubation for 20 min at 56 8C in the SDS-sample buffer (without mercaptoethanol) and analyzed with SDS-PAGE after boiling in the presence of mercaptoethanol using an 11% acrylamide gel (B). Dots represent the bands of proteins specifically immunoprecipitated using anti-F3 (F), anti-NAP-22 (N), anti-GAP-43 (A), and anti-Go a (G). The immunoreactivity of these bands was confirmed with Western blotting.
functional lipids, but also a factory metabolizing these lipids. Interestingly, the phosphatidylinositol 4-phosphate is an activator of the Cl2-pump ATPase [5,17]. Further studies focused on the distribution and regulation of the Cl2-pump will be interesting not only to understand the physiological regulatory mechanisms of neurons but also to unravel the physiological function of raft.
Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas (15013237) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by Natural Sciences Research Assistance from Asahi Glass Foundation. We thank Dr Chiyoko Inagaki (Kansai Medical University, Osaka, Japan) for valuable discussions and suggestions.
References
Fig. 2. Fractionation of the ATPase activities during the raft preparation. The 100,000 £ g pellet fraction (P3) was separated into Brij 96-soluble (BrS), Brij 96-insoluble high-density (Br-HDM), and Brij 96-insoluble lowdensity (Br-LDM) fractions. The ATPase activity was measured in the presence or absence of EAA (0.3 mM) as described in Fig. 1. Bars represent SD (n ¼ 4). *P , 0:05.
[1] R.G.W. Anderson, The caveola membrane system, Annu. Rev. Biochem. 67 (1998) 199–225. [2] B. Barylko, S.H. Gerber, D.D. Binns, N. Grichine, M. Khvotchev, T.C. Sudhof, J.P. Albanesi, A novel family of phosphatidylinositol 4kinases conserved from yeast to humans, J. Biol. Chem. 276 (2001) 7705–7708. [3] D.A. Brown, E. London, Structure and function of sphingolipid- and cholesterol-rich membrane rafts, J. Biol. Chem. 275 (2000) 17221–17224. [4] T.D. Foley, The lipid peroxidation product, 4-hydroxynonenal, potently and selectively inhibits synaptic plasma membrane ecto-
S. Maekawa, K. Taguchi / Neuroscience Letters 362 (2004) 158–161
[5]
[6]
[7] [8]
[9]
[10]
[11]
ATPase activity, a putative regulator of synaptic ATP and adenosine, Neurochem. Res. 24 (1999) 1241–1248. N. Hattori, K. Kitagawa, T. Higashida, K. Yagyu, S. Shimohama, T. Wataya, G. Perry, M.A. Smith, C. Inagaki, Cl2-ATPase and Naþ, KþATPase activities in Alzheimer’s disease brains, Neurosci. Lett. 254 (1998) 141–144. S. Iino, S. Kobayashi, S. Makeawa, Immunocytochemical localization of a novel acidic calmodulin binding protein, NAP-22, in the rat brain, Neuroscience 91 (1999) 1435–1444. C. Inagaki, N. Hattori, K. Kitagawa, X.-T. Zeng, K. Yagyu, Cl2ATPase in rat brain and kidney, J. Exp. Zool. 289 (2001) 224– 231. D.H. Jones, A.I. Matus, Isolation of synaptic plasma membrane from brain by combined flotation-sedimentation density gradient centrifugation, Biochim. Biophys. Acta 356 (1974) 276–287. S. Maekawa, S. Iino, S. Miyata, Molecular characterization of the detergent-insoluble cholesterol-rich membrane microdomain (raft) of the central nervous system, Biochim. Biophys. Acta 1610 (2003) 261–270. S. Maekawa, H. Kumanogoh, N. Funatsu, N. Takei, K. Inoue, Y. Endo, K. Hamada, Y. Sokawa, Identification of NAP-22 and GAP-43 (neuromodulin) as major protein components in a Triton insoluble low density fraction of rat brain, Biochim. Biophys. Acta 1323 (1997) 1–5. S. Maekawa, M. Maekawa, S. Hattori, S. Nakamura, Purification and molecular cloning of a novel acidic calmodulin-binding protein from rat brain, J. Biol. Chem. 268 (1993) 13703–13709.
161
[12] S. Maekawa, C. Sato, K. Kitajima, N. Funatsu, H. Kumanogoh, Y. Sokawa, Cholesterol-dependent localization of NAP-22 on a neuronal membrane microdomain (raft), J. Biol. Chem. 274 (1999) 21369– 21374. [13] M. Masserini, P. Palestini, M. Pitto, Glycolipid-enriched caveolae and caveolae-like domains in the nervous system, J. Neurochem. 73 (1999) 1–11. [14] J.T. Peniston, A.G. Filoteo, C.S. McDonough, E. Carafoli, Purification, reconstitution, and regulation of plasma membrane Ca2þpumps, Methods Enzymol. 157 (1988) 340–360. [15] L. Plesner, Ecto-ATPases: identities and functions, Int. Rev. Cytol. 158 (1995) 140 –214. [16] T. Shimizu, Y.Y. Toyoshima, R.D. Vale, Use of ATP analogs in motor assays, Methods Cell Biol. 39 (1993) 167–177. [17] T. Shiroya, R. Fukunaga, K. Akashi, N. Shimada, Y. Takagi, T. Nishino, M. Hara, C. Inagaki, An ATP-driven Cl2 pump in the brain, J. Biol. Chem. 265 (1989) 17416–17421. [18] K. Simons, E. Ikonen, Functional rafts in cell membranes, Nature 387 (1997) 569–572. [19] Y. Yamamoto, Y. Sokawa, S. Maekawa, Biochemical evidence for the presence of NAP-22, a novel acidic calmodulin binding protein, in the synaptic vesicles of rat brain, Neurosci. Lett. 224 (1997) 127 –130. [20] M. Zabe, W.L. Dean, Plasma membrane Ca2þ-ATPase associates with the cytoskeleton in activated platelets through a PDZ-binding domain, J. Biol. Chem. 276 (2001) 14704–14709.
Localization of the Cl2-ATPase activity on NAP-22 enriched membrane microdomain (raft) of rat brain Shohei Maekawa*, Katsutoshi Taguchi Division of Bioinformation, Department of Biosystems Science, Graduate School of Science and Technology, Kobe University, Kobe 657-8501, Japan Received 25 February 2004; received in revised form 11 March 2004; accepted 12 March 2004
Abstract Much attention has been paid to the membrane microdomain enriched in cholesterol and sphingolipids called raft. In the central nervous system, however, the physiological role of this domain is not so evident at present, partly because of the complexity of the protein components in the raft fraction. In this study we surveyed ATPase activities in the raft fraction obtained from the synaptic plasma membrane of rat brain and found the enrichment of an ethacrynic acid-sensitive ATPase (Cl2-pump) activity. Immunoprecipitation experiments using antibodies to raft-localized proteins showed the co-precipitation of the ATPase activity with NAP-22, a major raft-localized protein. This result suggests the participation of the raft in the regulation of ion transport in addition to the presence of heterogeneity of raft domains in neurons. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Cl2-ATPase; Raft; Neuron; NAP-22; Cholesterol
Lateral heterogeneities in the classical fluid-mosaic model of cell membranes are now envisaged as caveloae-like microdomains or ‘rafts’, that are enriched in cholesterol, glycosphingolipids, and specific membrane proteins including glycosylphosphatidylinositol (GPI)-anchored proteins [1,3,9,18]. Since many signal-transducing molecules such as trimeric G proteins, protein tyrosine kinases, cytoskeletal proteins, and calmodulin binding proteins are present in this region, raft is considered as a signal-transducing region of the cell [1,9 – 13]. From the preferential transport of GPIanchored proteins to the apical region of the epithelial cells, the raft is also considered to be important to the establishment and maintenance of cell polarity [18]. Elucidation of the molecular interaction in this region is, hence, important not only to understand the molecular mechanisms of signal transduction in cells but also to understand the intracellular sorting mechanisms of this region. Raft is generally described as a Triton-insoluble fraction of low density, i.e. a fraction recovered in a low buoyant density fraction after Triton treatment and sucrose density gradient centrifugation. During this procedure, Triton-insoluble lipid domains are thought to assemble to make vesicles, and proteins having affinities to these lipids are recovered in this * Corresponding author. Tel./fax: þ81-78-803-6507. E-mail address: [email protected] (S. Maekawa).
fraction. In the brain-derived raft, specific localization of GPI-anchored proteins, Src family protein kinases, trimeric G proteins, and neuron-specific calmodulin binding proteins (NAP-22 and GAP-43) was reported [9,10,13]. Among these proteins, NAP-22 is very important in considering the organization of the neuronal raft domains, because this protein is prominent in amount and is neuron-specific, has a cholesterol binding activity, and induces a cholesterol enriched domain in vitro and in vivo [9,13]. In the cell membrane many energy-dependent transporters perform their functions to sustain the cellular function. One of the main transporters is the Naþ, Kþ-ATPase, which maintains the gradients of Naþ and Kþ ions between the cell membrane and is sensitive to ouabain. The plasma membrane Ca2þ-pump is another ATPase which excludes Ca2þ ions and the activity is enhanced by Ca2þ-calmodulin [14]. In addition to these proteins, Inagaki and her colleagues identified a Cl2-pump as an ethacrynic acid (EAA)-sensitive ATPase [5,7,17]. The Cl2-pump and a brain type Kþ/Cl2 cotransporter are known to develop an inwardly directed Cl2 gradient which enables hyperpolarizing inhibitory postsynaptic potentials generated by transmitter-operated Cl2 channel opening [5,7,17]. In order to know the physiological contribution of raft on the cell function, the ATPase activities in the raft fraction
0304-3940/03/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.03.020
S. Maekawa, K. Taguchi / Neuroscience Letters 362 (2004) 158–161
were surveyed. The presence of the ouabain-sensitive ATPase and an enrichment of EAA-sensitive ATPase activity were observed. The preparation of the synaptic plasma membrane fraction (SPM) from 6-week-old rat (Wistar) brains and the detergent-insoluble low-density membrane fraction (raft) was prepared from SPM using 1% Triton X-100 (Tx-LDM) or 0.5% Brij 96 (Br-LDM) as described previously [8,19]. In order to measure the activity distribution during tissue fractionation, rat forebrain (3week-old) was homogenized in 8 vol of 0.25 M sucrose, 10 mM Tris –HCl, 1 mM EDTA, and 1 mM phenylmethylsulfonyl fluoride (PMSF) (pH 7.4). After centrifugation at 1000 £ g for 10 min, the supernatant was saved. The pellet fraction (P1) was homogenized again in 4 vol of the above solution and re-centrifuged. The recovered supernatants were mixed and centrifuged at 10,000 £ g for 20 min. The pellet fraction (P2) was saved and the supernatant was further centrifuged at 100,000 £ g for 30 min to recover the pellet fraction (P3). The sensitivity of the ATPase activity to EAA was most prominent in the P3 fraction as described previously [5]. The low-density detergent-insoluble membrane fraction (raft), the high-density detergent-insoluble fraction (HDM), and the detergent soluble fraction (BS) were hence prepared from the P3 fraction using 0.5% Brij 96 as described previously [12]. The raft fraction was prepared from the P3 fraction and further extracted with 0.5% MEGA10 (w/v) in 10 mM Tris– HCl, 0.14 M NaCl, 1 mM MgCl2, and 0.2 mM EGTA (pH 7.4) at 37 8C for 20 min. After centrifugation at 100,000 £ g for 30 min, the supernatant was recovered. The supernatant was incubated for 2 h at 4 8C with CNBr-activated Sepharose CL4B beads on which several monoclonal antibodies were coupled. After washing the beads three times with a solution described above (without MEGA10), the beads were washed once with a medium for the ATPase assay (described below) without ATP. The ATPase activity was then measured with the addition of ATP or the bound proteins were eluted through incubation with 1 vol of 2 £ SDS-sample buffer (without mercaptoethanol) at 56 8C for 20 min. The Naþ, Kþ-ATPase was assayed in 10 mM Tris–HCl (pH 7.4), 5 mM MgCl2, 1 mM EDTA, 0.1 M NaCl, 10 mM KCl, 0.5 mM ATP, and 0.02 mg/ml protein, with or without 1 mM ouabain at 37 8C. Naþ, Kþ-ATPase activity was calculated by subtracting the activity in the presence of 1 mM ouabain from the total ATPase activity. The difference between the ouabain-insensitive activities in the presence or absence of 0.3 mM EAA was designated as the Cl2-ATPase activity [5,17]. The liberated phosphate was measured to assay the ATPase activity with the malachite green method [16]. SDS-PAGE analysis, Western blotting, and protein determination were performed as described previously [10 – 12]. Production of monoclonal antibodies against NAP-22, GAP-43, and F3 was described previously [9,10, 12]. At first, we assayed the hydrolysis activities of Triton-
159
raft and Brij-raft using ATP, GTP, ADP, and AMP as the substrates. Both rafts showed high activities to hydrolyze ATP and AMP. Since brain contains the GPI-anchored form of 50 -nucleotidase, the AMPase activity observed here could be ascribed to this enzyme [1,4,13]. In order to address the molecular species of raft-localized ATPases, the effects of various specific inhibitors were studied. Bafilomycin A1 (10 mM, V-ATPase), 2,3-butanedione-2-monoxime (10 mM, myosin ATPase), EGTA (2 mM, Ca2þ-ATPase), and 4hydroxynonenal (50 mM, ecto-ATPase) showed little effects on the activity (data not shown) [4,14 – 16,20]. On the other hand, ethacrynic acid (0.3 mM) and ouabain (1 mM) inhibited the activity to some extent. Fig. 1 shows the effect of EAA on SPM and SPM-derived rafts. The ATPase activity of Brij-raft was reduced to about 60% in the presence of EAA. The inhibitory effect of EAA was also observed in Triton-raft, although the effect was not so high. In contrast, EAA showed little effect on the ATPase activity of SPM. This result shows the enrichment of the EAA-sensitive (Cl2-pump) activity in the raft fraction. The ouabain-sensitive ATPase activity, in contrast, showed no specific localization on raft and further Western blotting using an antibody to this enzyme confirmed this result (data not shown). In order to analyze the localization of the EAA-sensitive ATPase in raft, the P3 fraction was further fractionated using Brij 96 to obtain Brij-soluble fraction (BrS), raft (Br-LDM), and Brij-insoluble high-density fraction (Br-HDM), and the ATPase activities were measured. The ouabain-sensitive ATPase activity was observed in both the BrS and the B-LDM fraction (data not shown). The EAA-sensitive ATPase was, in contrast, largely recovered in the Br-LDM (raft) fraction. The Br-HDM showed a substantial ATPase activity. This activity, however, showed poor EAA sensitivity (Fig. 2). Inagaki and her colleagues showed the solubilization of Cl2-pump with a detergent, MEGA10 [5,7]. Since the solubilized activity was recovered in a high molecular weight range in a native gel
Fig. 1. Presence of an EAA-sensitive ATPase in the raft (LDM) fraction obtained from SPM fraction using Triton X-100 or Brij 96. The ATPase activity was assayed in 10 mM Tris –HCl, 5 mM MgCl2, 1 mM EDTA, 0.1 M NaCl, 10 mM KCl, 0.5 mM ATP, 1 mM ouabain, and 0.02 mg/ml protein at 37 8C for 10 min in the presence or absence of 0.3 mM EAA. Bars represent SD (n ¼ 4). *P , 0:05.
160
S. Maekawa, K. Taguchi / Neuroscience Letters 362 (2004) 158–161
electrophoresis, they suggested an oligomeric nature of the enzyme [7]. The result described here suggests the association of the enzyme in raft in a multimolecular complex. The molecular characterization of brain-derived raft showed the localization of F3, NAP-22, GAP-43, and trimeric G protein Go as the major raft components [9,10]. Since the presence of a variety of rafts with different protein and lipid components is known, further studies on the colocalization the Cl2-pump with these proteins is of great importance. Immunoprecipitation experiments using monoclonal antibodies to these major proteins were, hence, applied after the partial solubilization of the raft with MEGA10. The SDS-PAGE analysis of the immunoprecipitated samples showed the specific recovery of the antigens but not other major proteins (Fig. 3B). This result shows that the raft proteins solubilized with MEGA10 are separable from each other. The ATPase analysis of these fractions showed a recovery of the EAA-sensitive ATPase activity in the NAP-22 immunoprecipitate but not in the other immunoprecipitates (Fig. 3A). Since raft is a lipid-based structure and the detergent only partially breaks down the raft through partial extraction of lipids, the recovery of the ATPase in a NAP-22 enriched raft does not directly mean an interaction between NAP-22 and the ATPase. Co-precipitation, however, suggests the possibility that these proteins localize in the same raft within the neuron. Since the molecular cloning of the Cl2-pump has not yet been done, the precise comparison of the localization pattern of the Cl2-pump with that of NAP-22 is impossible at present [6]. A rapid increase of the Cl2-ATPase activity during the neonatal stages is observed and this pattern is very similar to that of NAP-22 [7,9,17]. The localization of phosphatidylinositol 40 ,50 -bisphosphate (PIP2) in raft is well recognized and recent studies showed the localization of a novel type of phosphatidylinositol 4-kinase (PI-4K II) in the raft fraction [1,2,18]. The raft is therefore not only the storehouse of
Fig. 3. Immunoprecipitation of MEGA10 solubilized raft fraction using monoclonal antibodies to NAP-22 (NAP), GAP-43 (GAP), F3 (F3), and Go a subunit (Go). The antibody coupled gels were mixed with the fraction at 4 8C for 2 h. After washing the ATPase activity of the gels was assayed (A) in the presence or absence of EAA. Bars represent SD (n ¼ 4). *P , 0:05. The bound proteins were eluted from the gels with incubation for 20 min at 56 8C in the SDS-sample buffer (without mercaptoethanol) and analyzed with SDS-PAGE after boiling in the presence of mercaptoethanol using an 11% acrylamide gel (B). Dots represent the bands of proteins specifically immunoprecipitated using anti-F3 (F), anti-NAP-22 (N), anti-GAP-43 (A), and anti-Go a (G). The immunoreactivity of these bands was confirmed with Western blotting.
functional lipids, but also a factory metabolizing these lipids. Interestingly, the phosphatidylinositol 4-phosphate is an activator of the Cl2-pump ATPase [5,17]. Further studies focused on the distribution and regulation of the Cl2-pump will be interesting not only to understand the physiological regulatory mechanisms of neurons but also to unravel the physiological function of raft.
Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas (15013237) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by Natural Sciences Research Assistance from Asahi Glass Foundation. We thank Dr Chiyoko Inagaki (Kansai Medical University, Osaka, Japan) for valuable discussions and suggestions.
References
Fig. 2. Fractionation of the ATPase activities during the raft preparation. The 100,000 £ g pellet fraction (P3) was separated into Brij 96-soluble (BrS), Brij 96-insoluble high-density (Br-HDM), and Brij 96-insoluble lowdensity (Br-LDM) fractions. The ATPase activity was measured in the presence or absence of EAA (0.3 mM) as described in Fig. 1. Bars represent SD (n ¼ 4). *P , 0:05.
[1] R.G.W. Anderson, The caveola membrane system, Annu. Rev. Biochem. 67 (1998) 199–225. [2] B. Barylko, S.H. Gerber, D.D. Binns, N. Grichine, M. Khvotchev, T.C. Sudhof, J.P. Albanesi, A novel family of phosphatidylinositol 4kinases conserved from yeast to humans, J. Biol. Chem. 276 (2001) 7705–7708. [3] D.A. Brown, E. London, Structure and function of sphingolipid- and cholesterol-rich membrane rafts, J. Biol. Chem. 275 (2000) 17221–17224. [4] T.D. Foley, The lipid peroxidation product, 4-hydroxynonenal, potently and selectively inhibits synaptic plasma membrane ecto-
S. Maekawa, K. Taguchi / Neuroscience Letters 362 (2004) 158–161
[5]
[6]
[7] [8]
[9]
[10]
[11]
ATPase activity, a putative regulator of synaptic ATP and adenosine, Neurochem. Res. 24 (1999) 1241–1248. N. Hattori, K. Kitagawa, T. Higashida, K. Yagyu, S. Shimohama, T. Wataya, G. Perry, M.A. Smith, C. Inagaki, Cl2-ATPase and Naþ, KþATPase activities in Alzheimer’s disease brains, Neurosci. Lett. 254 (1998) 141–144. S. Iino, S. Kobayashi, S. Makeawa, Immunocytochemical localization of a novel acidic calmodulin binding protein, NAP-22, in the rat brain, Neuroscience 91 (1999) 1435–1444. C. Inagaki, N. Hattori, K. Kitagawa, X.-T. Zeng, K. Yagyu, Cl2ATPase in rat brain and kidney, J. Exp. Zool. 289 (2001) 224– 231. D.H. Jones, A.I. Matus, Isolation of synaptic plasma membrane from brain by combined flotation-sedimentation density gradient centrifugation, Biochim. Biophys. Acta 356 (1974) 276–287. S. Maekawa, S. Iino, S. Miyata, Molecular characterization of the detergent-insoluble cholesterol-rich membrane microdomain (raft) of the central nervous system, Biochim. Biophys. Acta 1610 (2003) 261–270. S. Maekawa, H. Kumanogoh, N. Funatsu, N. Takei, K. Inoue, Y. Endo, K. Hamada, Y. Sokawa, Identification of NAP-22 and GAP-43 (neuromodulin) as major protein components in a Triton insoluble low density fraction of rat brain, Biochim. Biophys. Acta 1323 (1997) 1–5. S. Maekawa, M. Maekawa, S. Hattori, S. Nakamura, Purification and molecular cloning of a novel acidic calmodulin-binding protein from rat brain, J. Biol. Chem. 268 (1993) 13703–13709.
161
[12] S. Maekawa, C. Sato, K. Kitajima, N. Funatsu, H. Kumanogoh, Y. Sokawa, Cholesterol-dependent localization of NAP-22 on a neuronal membrane microdomain (raft), J. Biol. Chem. 274 (1999) 21369– 21374. [13] M. Masserini, P. Palestini, M. Pitto, Glycolipid-enriched caveolae and caveolae-like domains in the nervous system, J. Neurochem. 73 (1999) 1–11. [14] J.T. Peniston, A.G. Filoteo, C.S. McDonough, E. Carafoli, Purification, reconstitution, and regulation of plasma membrane Ca2þpumps, Methods Enzymol. 157 (1988) 340–360. [15] L. Plesner, Ecto-ATPases: identities and functions, Int. Rev. Cytol. 158 (1995) 140 –214. [16] T. Shimizu, Y.Y. Toyoshima, R.D. Vale, Use of ATP analogs in motor assays, Methods Cell Biol. 39 (1993) 167–177. [17] T. Shiroya, R. Fukunaga, K. Akashi, N. Shimada, Y. Takagi, T. Nishino, M. Hara, C. Inagaki, An ATP-driven Cl2 pump in the brain, J. Biol. Chem. 265 (1989) 17416–17421. [18] K. Simons, E. Ikonen, Functional rafts in cell membranes, Nature 387 (1997) 569–572. [19] Y. Yamamoto, Y. Sokawa, S. Maekawa, Biochemical evidence for the presence of NAP-22, a novel acidic calmodulin binding protein, in the synaptic vesicles of rat brain, Neurosci. Lett. 224 (1997) 127 –130. [20] M. Zabe, W.L. Dean, Plasma membrane Ca2þ-ATPase associates with the cytoskeleton in activated platelets through a PDZ-binding domain, J. Biol. Chem. 276 (2001) 14704–14709.