- Email: [email protected]
Brain-derived neurotrophic factor regulates the expression of AMPA receptor proteins in neocortical neurons
Pergamon
PII:
Neuroscience Vol. 88, No. 4, pp. 1009–1014, 1999 Copyright 1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/99 $19.00+0.00 S0306-4522(98)00496-5
Letter to Neuroscience BRAIN-DERIVED NEUROTROPHIC FACTOR REGULATES THE EXPRESSION OF AMPA RECEPTOR PROTEINS IN NEOCORTICAL NEURONS M. NARISAWA-SAITO,*‡ J. CARNAHAN,† K. ARAKI,* T. YAMAGUCHI* and H. NAWA*‡ *Department of Molecular Neurobiology, Brain Research Institute, Niigata University, Niigata 951, Japan †Amgen Center, Thousand Oaks, CA 91320, U.S.A. Key words: neurotrophin, NMDA receptor, synaptic plasticity, glutamate receptor.
The role of the neurotrophins; nerve growth factor,22 brain-derived neurotrophic factor,2 neurotrophin-316,24 and neurotrophin-4/5,3 in synaptic development and plasticity has been extensively investigated.23,33 The neurotrophins regulate synaptic transmission1,13,19,34 as well as neural development4,5,12,18,25 in the brain. However, the mechanisms underlying these processes are unknown. In this study we show that brain-derived neurotrophic factor triggers an increase in á-amino-3hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)type glutamate receptor (GluR) proteins without significant changes in their messenger RNA levels. Brain-derived neurotrophic factor treatment specifically increased the protein levels of GluR1 (19322%) and GluR2/3 (18211%) in cultured rat neocortical neurons. In contrast, nerve growth factor and neurotrophin-3 failed to alter the protein levels of these neurons, and brain-derived neurotrophic factor effects on N-methyl-D-aspartate-type glutamate receptors were either modest or negligible. Immunocytochemical studies indicated that the increase in AMPA receptor proteins reflects the induction of their neuronal expression, but not selective neuronal survival. In agreement with these results, cortical neurons from brain-derived neurotrophic factor-knockout mice exhibited a reduction in AMPA receptor proteins in the cytoskeletal fraction containing postsynaptic proteins. Thus, the neurotrophin plays a crucial role in modulating the expression of AMPA receptors presumably ‡Authors to whom correspondence should be addressed. Abbreviations: AMPA, á-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid; BDNF, brain-derived neurotrophic factor; áCAMKII, á-calmodulin-dependent kinase II; GluR, glutamate receptor; HEPES, N-2hydroxyethylpiperazine-N -2-ethanesulfonic acid; NGF, nerve growth factor; NMDA, N-methyl--aspartate; NT, neurotrophin; PSD, postsynaptic density; SDS, sodium dodecyl sulfate.
at translational or post-translation levels and is implicated in synaptic development and plasticity. 1998 IBRO. Published by Elsevier Science Ltd. We prepared primary cultures of neocortical neurons from fetal rats and grew them in serum-free medium with daily applications of neurotrophins (50 ng/ml). The serum-free culture condition prevented the proliferation of non-neuronal cells so that they accounted for less than 5% of the total number of cells.27 Applications of brain-derived neurotrophic factor (BDNF) markedly increased the protein expression of GluR1 and GluR2/3 receptors. By setting the average control receptor level as 100 for each immunoblot, their protein levels were found to change from 10012% to 19322% for GluR1, and from 1006% to 18211% for GluR2/3 (n=3; P<0.05). A typical immunoblot is shown in Fig. 1. In contrast, the other neurotrophins, nerve growth factor (NGF) and neurotrophin-3, failed to alter the protein levels of the AMPA receptors. Moreover, expressions of the N-methyl--aspartate (NMDA) receptor 1 and 2A/B subunits were not significantly affected: control levels and the BDNF-stimulated levels were 1005% and 1019% for NMDA receptor-1 (NMDAR1) protein, and 1008% and 1205% for NMDAR2A/B protein, respectively. In addition, other cytoskeletal proteins, á-tubulin and neurofilament protein, were not affected. Importantly, K252b (1 µM), an inhibitor of receptor tyrosine kinases, blocked the effects of BDNF. Thus, these results suggest that the BDNF activity is quite specific for the AMPA receptors. To analyse the effect of BDNF on individual neurons, neocortical cultures were immunostained with anti-GluR1 and anti-GluR2/3 antibodies (Fig. 2). Although the variation in staining intensity was substantial, about half of the neurons in control
1009
1010
M. Narisawa-Saito et al.
Fig. 1. Effects of neurotrophins on glutamate receptor proteins in cultured rat neocortical neurons. Proteins were extracted from embryonic neocortical neurons grown for five days either with or without () daily applications of 50 ng/ml NT-3, BDNF and NGF, and BDNF plus 1 µM K252b. The protein was separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, transferred to a membrane, and probed with antibodies directed against GluR1, GluR2/3, NMDAR1, NMDAR2A/B (all recognize their carboxyl termini; Chemicon), á-tubulin (SANBIO) and anti-neurofilaments (NF) 200/150,000 mol. wt (a gift from Dr Takahashi, Niigata University, Japan). Immunoreactivity was visualized by chemiluminescence (ECL kit; Amersham). Whole cerebral neocortices of embryonic day-18 rats were mechanically dissociated and plated onto laminin-coated dishes. Cortical neurons were grown with Dulbecco’s modified Eagle’s medium containing 0.5 mM purified glutamine (Ajinomoto Co), nutrient mixture N2 and 10 mM HEPES (pH 7.3).27 The mouse 2.5S NGF was obtained from Boehringer Mannheim, and BDNF and NT-3 were from Amgen (Thousand Oaks, CA, U.S.A.).
culture exhibited a moderate level of GluR1- or GluR2/3-immunoreactivity (Table 1), which is almost consistent with previous in vivo immunohistochemical observations.8 Cells with strong immunoreactivity in their perikarya were only rarely seen. After growth
with BDNF, however, the frequency of neocortical neurons that contained strong immunoreactivity for either GluR1 or GluR2/3 more markedly increased than that exhibiting the moderate immunoreactivity for the receptors. Total cell densities were not influenced by BDNF, suggesting that BDNF did not influence neuronal survival under the culture conditions used.27 To test whether BDNF enhances the mRNA expression of the AMPA receptors, we measured their GluR mRNA levels by RNA blotting (Fig. 3). Interestingly, the AMPA receptor mRNA levels were not significantly altered by treatment with BDNF: mRNA changes over control cultures were calculated as 995% for GluR1 mRNA, 955% for GluR2 mRNA and 9910% GluR3 mRNA when cyclophilin mRNA was used as an internal control (n=3). We also carried out the RNase protection assay and confirmed the equality in GluR1 mRNA: GluR1 mRNA levels were 93.45.9% in the BDNF-treated culture in comparison with 1002.5% in control culture (n=4). Again, these results suggest that the regulation of AMPA receptor expression mainly involves modulating translational and/or post-translational mechanisms. Analyses of BDNF-knockout mice have confirmed the idea that BDNF is essential for normal neural development12,18 and synaptic transmission.21,28 We also examined BDNF-knockout mice to study the effect of BDNF on AMPA receptor proteins in vivo. The neocortical GluR levels of the BDNF knockout mice exhibited decreasing tendency compared to those of wild mice but the difference did not reach statistical significance in total protein fractions (data not shown). Thus, we fractionated the subcellular components of the neocortical tissue into membrane (S), postsynaptic density/cytoskeletal (P) and cytoplasmic fractions according to the method of McNeill and Colbran26 (Fig. 4A). More than 90% of GluR2/3 immunoreactivity was recovered in the S or P fraction. Between the S and P fractions, immunoreactivity for á-calmodulin-dependent kinase II (áCAMKII), a marker for postsynaptic density (PSD),20 was fractionated into the P fraction with high efficiency (Table 2). The BDNF-knockout mice exhibited a marked reduction in GluR2/3 immunoreactivity (P<0.05) in the P fraction, but not in the S fraction (Fig. 4B, Table 2). In contrast, levels of áCAMKII in the P fractions of both genotypes were almost same. These observations suggest that endogenous BDNF specifically influences the subcellular expression of the AMPA receptors in vivo. DISCUSSION
In the present study, our results suggest that BDNF can specifically increase protein levels of AMPA receptors. Reversibly, expression of BDNF itself is known to be induced by activation of AMPA
AMPA receptor phenotype regulated by BDNF
1011
Fig. 2. Immunocytochemical analysis of the BDNF effects on AMPA receptors. Cultures of rat neocortical neurons were prepared similarly and grown for five days with or without BDNF. Paraformaldehyde-fixed cells were processed for immunostaining with anti-AMPA receptor (GluR1 and GluR2/3) antibodies as well as with anti-NMDA receptor (NMDAR1 and NMDAR2A/B) antibodies. Antigen–antibody complexes were visualized by the ABC method, with nickel enhancement.27 Scale bar=100 µm.
receptors in the brain.11,35 Therefore, these two observations indicate that BDNF and AMPA receptors mutually interact with each other to regulate positively the other expression. Their interactions might contribute to normal brain development and plasticity.23,33 The receptor increase appeared to occur within only a part of neuronal populations because some neurons showed no response to BDNF. We have previously described a similar phenotypic regulation of neocortical neurons by neurotrophins: BDNF upregulates peptidergic expression in neocortical and striatal neurons in vivo and in vitro without changing their survival rate.10,27 All these phenotypes are associated with nonpyramidal GABAergic neurons
and are impaired in the neocortex of BDNFknockout mice.18 Interestingly, most of the neocortical neurons that show strong GluR1 immunoreactivity are also GABAergic nonpyramidal neurons,8 some of which express higher levels of BDNF receptor (trkB) mRNA as well.6 These results suggest that BDNF may act mainly on this GABAergic neuronal population to result in an increase in the AMPA receptor proteins. Our RNA analyses indicated that the actions of BDNF on the receptor expression presumably involve translational or post-translational processes. This BDNF activity appears to be different from the reported activities of other neurotrophic factors, such as Neu differentiation factor which induces mRNA
1012
M. Narisawa-Saito et al. Table 1. Increased frequency of AMPA-receptor immunoreactive neurons in culture GluR1 Positive
GluR2/3 Positive
Treatment
Moderate
Strong
Moderate
Strong
Cell density (cells/mm2)
None BDNF
47.82.9 43.72.3
4.20.5 33.53.2**
46.31.4 59.22.1**
6.20.8 12.51.9*
75744 85645
Six sister cultures of neocortical neurons were prepared as described in Fig. 2. Since the intensity of positive immunoreactivity was heterogenous among neurons, immunoreactive cells were classified into two groups (cells with moderate staining and those with strong staining) and their numbers were counted. The frequency of these positive cells (%) was calculated and presented. Statistical analysis was performed with a one-way analysis of variance (ANOVA). Results are expressed as meansS.E.M. (n=6). *P<0.05, **P<0.01 compared with control. Note: cell densities were not significantly altered by BDNF.
Fig. 3. BDNF effects on mRNA levels for the AMPA receptors. RNA was separated using a 1% agarose gel, transferred to a membrane, and probed with 32P-labeled cDNA for GluR1, GluR2 and GluR3.17 The RNA extracted from rat whole brains was used as a positive control. A typical example of a RNA blot is shown for display. 32P-labeled cDNA probes for rat GluR1 (a 2.2-kb fragment of p59/2; a gift from Dr Heinemann, The Salk Institute, CA, U.S.A.), mouse GluR2 (a 1.5-kb fragment; a gift from Dr Sakimura, Niigata University, Japan), rat GluR3 (a 269nt–1337nt fragment produced by reverse transcriptionpolymerase chain reaction) and rat cyclophilin (a 0.7-kb fragment) were generated using the Random Primed DNA Labeling kit (Boehringer). These probes were sequentially hybridized to three RNA filters (n=3) and the relative signal intensity of GluR mRNAs to that of cyclophilin mRNA was determined. Alternatively, an RNase protection assay was performed by using the RNase protection kit (Boehringer Mannheim) with a 367-base riboprobe that was made from T7 promoter to a BalI site of the rat GluR1 plasmid (data not shown).
for neurotransmitter receptors.30 Post-translational modifications of transmitter receptors at the neuromuscular synapse have been extensively studied. The non-receptor-type tyrosine kinase, Src, and musclespecific kinase are associated with nicotinic acetylcholine receptors, and their activation has been implicated in receptor clustering and stabilization at postsynaptic sites by altering their cytoskeletal inter-
actions.14,15,32 The BDNF receptor trkB and AMPA receptors are all intrinsic components of PSD.9,29,31 Thus, it is quite natural to speculate that the protein phosphorylation cascade, starting with the activation of the trkB kinase, may contribute to the regulation of AMPA receptor stability in the brain. However, we cannot deny a possibility that BDNF might also affect gene transcription of the receptors whose
AMPA receptor phenotype regulated by BDNF
1013
Table 2. Subcellular fractionation of synaptic proteins from the neocortex of wild-type and brain-derived neurotrophic factor-knockout mice. Protein yield Molecule áCAMKII SNAP25 GluR2/3
Fig. 4. The expression of GluR2/3 protein in the subcellular fractions from the neocortex of BDNF-knockout mice. (A) The subcellular fractionation carried out using the method of McNeill and Colbran.26 The neocortical homogenates were separated by centrifugation (100,000 g for 60 min) into the supernatant (cytosolic proteins) and the pellet containing crude membranes. The pellet was rehomogenized with 1% Triton X-100-containing buffer and centrifuged as before to give rise to supernatant 2 (S; membrane fraction) and precipitate 2 (P; PSD/cytoskeletal fraction). Authenticity of the fractionation was examined by western blotting of the protein markers, SNAP257 (a presynaptic membrane protein) and áCAMKII20 (a PSD protein) as shown in Table 2. Each lane contained 20 µg protein of the fraction. (B) BDNF-knockout mice12 were produced by crossing heterozygous mutant mice and grown until postnatal day 12–14. The subcellular proteins of each mouse neocortex were fractionated by the above method into S and P fractions, and protein levels of GluR2/3 in these fractions were similarly examined by western blotting. A window containing each pair of fractions represents the data from a wild mouse or a homozygous mouse (-/-) (n=3 animals for each genotype).
mRNA changes would be transient and disappear acutely. The neurotrophins can modulate basal synaptic transmission18,34 as well as long-term potentia-
Fraction
Wild (%)
S P S P
22.44.7 77.63.2** 86.510.4** 13.51.5
S P
5.00.9 1006.1
Homozygotes (%) 70.75.7
5.30.4 51.711.5*
Authenticity of the subcellular fractionation was ascertained by measuring the recoverey ratio of the marker proteins, SNAP25[7] and áCAMKII[20] (n=3) as shown in Fig. 4A. In addition, measuring the intensity of signals on western blots in Fig. 4B, levels of GluR2/3 protein as well as those of áCAMKII in the neocortex were compared between wild and BDNF-knockout mice (n=3). In the comparison for GluR2/3, the highest protein level was set as 100%. Statistical analysis was performed with a one-way ANOVA followed by the Bonferroni test. Results are expressed as meansS.E.M. **P<0.01 compared between the S and P fractions, and *P<0.05 compared between wild-type and the mutant mice. Note: protein levels of áCAMKII were similar in the P fraction of both genotypes.
tion.1,13 Changes in hippocampal neurotransmission induced by BDNF similarly depend upon translation, but not on transcription.19 Subsequent electrophysiological studies will illuminate how the total increase in the AMPA receptor expression is correlated with the acute BDNF-mediated changes of neurotransmission that were reported previously. In conclusion, the present results will provide novel insights into the mechanisms involved in the regulation of AMPA receptor expression during development and in synaptic plasticity. Acknowledgements—We thank Dr P. Ernfors and Dr R. Jaenisch for providing the brain tissues from BDNF mutant mice, Miss J. H. Kogan and Mrs C. Fujikawa for technical assistance, Dr S. Heinemann for rat GluR1 cDNA, Dr K. Sakimura for mouse GluR2 cDNA and GluR2/3 antibodies, and Dr T. Yamamoto for his advice. This project was supported by the Japan Society for the Promotion of Science (RFTF-96L00203) and Uehara Memorial Foundation.
REFERENCES
1.
Akaneya Y., Tsumoto T. and Hatanaka H. (1996) Brain-derived neurotrophic factor blocks long-term depression in rat visual cortex. J. Neurophysiol. 76, 4198–4201. 2. Barde Y. A., Edgar D. and Thoenen H. (1982) Purification of a new neurotrophic factor from mammalian brain. Eur. molec. Biol. Org. J. 1, 549–553. 3. Berkemeier L. R., Winslow J. W., Kaplan D. R., Nikolics K., Goeddel D. V. and Rosenthal A. (1991) Neurotrophin-5: a novel neurotrophic factor that activates trk and trkB. Neuron 7, 857–866. 4. Cabelli R. J., Shelton D. L., Segal R. A. and Shatz C. J. (1997) Blockade of endogenous ligands of trkB inhibits formation of ocular dominance columns. Neuron 19, 63–76. 5. Causing C. G., Gloster A., Aloyz R., Bamji S. X., Chang E., Fawcett J., Kuchel G. and Miller F. D. (1997) Synaptic innervation density is regulated by neuron-derived BDNF. Neuron 18, 257–267.
1014 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
M. Narisawa-Saito et al. Cellerino A., Maffei L. and Domenici L. (1996) The distribution of brain-derived neurotrophic factor and its receptor trkB in parvalbumin-containing neurons of the rat visual cortex. Eur. J. Neurosci. 8, 1190–1197. Chapman E. R., An S., Barton N. and Jahn R. (1994) SNAP-25, a t-SNARE which binds to both syntaxin and synaptobrevin via domains that may form coiled coils. J. biol. Chem. 269, 27,427–27,432. Conti F., Minell A. and Brecha N. C. (1994) Cellular localization and laminar distribution of AMPA glutamate receptor subunits mRNAs and proteins in the rat cerebral cortex. J. comp. Neurol. 350, 241–259. Craig A. M., Blackstone C. D., Huganir R. L. and Banker G. (1993) The distribution of glutamate receptors in cultured rat hippocampal neurons: postsynaptic clustering of AMPA-selective subunits. Neuron 10, 1055–1068. Croll S. D., Wiegand S. J., Anderson K. D., Lindsay R. M. and Nawa H. (1994) Regulation of neuropeptides in adult rat forebrain by the neurotrophins BDNF and NGF. Eur. J. Neurosci. 6, 1343–1353. Ernfors P., Bengzon J., Kokaia Z., Persson H. and Lindvall O. (1991) Increased levels of messenger RNAs for neurotrophic factors in the brain during kindling epileptogenesis. Neuron 7, 165–176. Ernfors P., Lee K. F. and Jaenisch R. (1994) Mice lacking brain-derived neurotrophic factor develop with sensory deficits. Nature 368, 147–150. Figurov A., Pozzo-Miller L. D., Olafsson P., Wang T. and Lu B. (1996) Regulation of synaptic responses to high-frequency stimulation and LTP by neurotrophins in the hippocampus. Nature 381, 706–709. Fuhrer C. and Hall Z. W. (1996) Functional interaction of Src family kinases with the acetylcholine receptor in C2 myotubes. J. biol. Chem. 271, 32,474–32,481. Fuhrer C., Sugiyama J. E., Taylor R. G. and Hall Z. W. (1997) Association of muscle-specific kinase MuSK with the acetylcholine receptor in mammalian muscle. Eur. molec. Biol. Org. J. 16, 4951–4960. Hohn A., Leibrock J., Bailey K. and Barde Y. A. (1990) Identification and characterization of a novel member of the nerve growth factor/brain-derived neurotrophic factor family. Nature 344, 339–341. Hollmann M. and Heinemann S. (1994) Cloned glutamate receptors. A. Rev. Neurosci. 17, 31–108. Jones K. R., Farinas I., Backus C. and Reichardt L. F. (1994) Targeted disruption of the BDNF gene perturbs brain and sensory neuron development but not motor neuron development. Cell 76, 989–999. Kang H. and Schuman E. M. (1996) A requirement for local protein synthesis in neurotrophin-induced hippocampal synaptic plasticity. Science 273, 1402–1406. Kennedy M. B. (1997) The postsynaptic density at glutamatergic synapses. Trends Neurosci. 20, 264–268. Korte M., Griesbeck O., Gravel C., Carroll P., Staiger V., Thoenen H. and Bonhoeffer T. (1996) Virus-mediated gene transfer into hippocampal CA1 region restores long-term potentiation in brain-derived neurotrophic factor mutant mice. Proc. natn. Acad. Sci. U.S.A. 93, 12,547–12,552. Levi-Montalcini R. (1987) The nerve growth factor: thirty-five years later. Eur. molec. Biol. Org. J. 6, 1145–1154. Lewin G. R. and Barde Y. A. (1996) Physiology of the neurotrophins. A. Rev. Neurosci. 19, 289–317. Maisonpierre P. C., Belluscio L., Squinto S., Ip N. Y., Furth M. E., Lindsay R. M. and Yancopoulos G. D. (1990) Neurotrophin-3: a neurotrophic factor related to NGF and BDNF. Science 247, 1446–1451. McAllister A. K., Katz L. C. and Lo D. C. (1997) Opposing roles for endogenous BDNF and NT-3 in regulating cortical dendritic growth. Neuron 18, 767–778. McNeill R. B. and Colbran R. J. (1995) Interaction of autophosphorylated Ca2+/calmodulin-dependent protein kinase II with neuronal cytoskeletal proteins. J. biol. Chem. 270, 10,043–10,049. Nawa H., Bessho Y., Carnahan J., Nakanishi S. and Mizuno K. (1993) Regulation of neuropeptide expression in cultured cerebral cortical neurons by brain-derived neurotrophic factor. J. Neurochem. 60, 772–775. Patterson S. L., Abel T., Deuel T. A. S., Martin K. C., Rose K. C. and Kandel E. C. (1996) Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron 16, 1137–1145. Petralia R. S. and Wenthold R. J. (1992) Light and electron immunocytochemical localization of AMPA-selective glutamate receptors in the rat brain. J. comp. Neurol. 318, 329–354. Sandrock A. W. Jr, Dryer S. E., Rosen K. M., Gozani S. N., Kramer R., Theill L. E. and Fischbach G. D. (1997) Maintenance of acetylcholine receptor number by neuregulins at the neuromuscular junction in vivo. Science 276, 599–603. Suen P. C., Wu K., Levine E. S., Mount H. T., Xu J. L., Lin S. Y. and Black I. B. (1997) Brain-derived neurotrophic factor rapidly enhances phosphorylation of the postsynaptic N-methyl--aspartate receptor subunit 1. Proc. natn. Acad. Sci. U.S.A. 94, 8191–8195. Swope S. L. and Huganir R. L. (1993) Molecular cloning of two abundant protein tyrosine kinases in Torpedo electric organ that associate with the acetylcholine receptor. J. biol. Chem. 268, 25,152–25,161. Thoenen H. (1995) Neurotrophins and neuronal plasticity. Science 270, 593–598. Tanaka T., Saito H. and Matsuki N. (1997) Inhibition of GABA-A synaptic responses by brain-derived neurotrophic factor (BDNF) in rat hippocampus. J. Neurosci. 17, 2959–2966. Zafra F., Hengerer B., Leibrock J., Thoenen H. and Lindholm D. (1990) Activity dependent regulation of BDNF and NGF mRNAs in the rat hippocampus is mediated by non-NMDA glutamate receptors. Eur. molec Biol. Org. J. 9, 3545–3550. (Accepted 21 August 1998)
PII:
Neuroscience Vol. 88, No. 4, pp. 1009–1014, 1999 Copyright 1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/99 $19.00+0.00 S0306-4522(98)00496-5
Letter to Neuroscience BRAIN-DERIVED NEUROTROPHIC FACTOR REGULATES THE EXPRESSION OF AMPA RECEPTOR PROTEINS IN NEOCORTICAL NEURONS M. NARISAWA-SAITO,*‡ J. CARNAHAN,† K. ARAKI,* T. YAMAGUCHI* and H. NAWA*‡ *Department of Molecular Neurobiology, Brain Research Institute, Niigata University, Niigata 951, Japan †Amgen Center, Thousand Oaks, CA 91320, U.S.A. Key words: neurotrophin, NMDA receptor, synaptic plasticity, glutamate receptor.
The role of the neurotrophins; nerve growth factor,22 brain-derived neurotrophic factor,2 neurotrophin-316,24 and neurotrophin-4/5,3 in synaptic development and plasticity has been extensively investigated.23,33 The neurotrophins regulate synaptic transmission1,13,19,34 as well as neural development4,5,12,18,25 in the brain. However, the mechanisms underlying these processes are unknown. In this study we show that brain-derived neurotrophic factor triggers an increase in á-amino-3hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)type glutamate receptor (GluR) proteins without significant changes in their messenger RNA levels. Brain-derived neurotrophic factor treatment specifically increased the protein levels of GluR1 (19322%) and GluR2/3 (18211%) in cultured rat neocortical neurons. In contrast, nerve growth factor and neurotrophin-3 failed to alter the protein levels of these neurons, and brain-derived neurotrophic factor effects on N-methyl-D-aspartate-type glutamate receptors were either modest or negligible. Immunocytochemical studies indicated that the increase in AMPA receptor proteins reflects the induction of their neuronal expression, but not selective neuronal survival. In agreement with these results, cortical neurons from brain-derived neurotrophic factor-knockout mice exhibited a reduction in AMPA receptor proteins in the cytoskeletal fraction containing postsynaptic proteins. Thus, the neurotrophin plays a crucial role in modulating the expression of AMPA receptors presumably ‡Authors to whom correspondence should be addressed. Abbreviations: AMPA, á-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid; BDNF, brain-derived neurotrophic factor; áCAMKII, á-calmodulin-dependent kinase II; GluR, glutamate receptor; HEPES, N-2hydroxyethylpiperazine-N -2-ethanesulfonic acid; NGF, nerve growth factor; NMDA, N-methyl--aspartate; NT, neurotrophin; PSD, postsynaptic density; SDS, sodium dodecyl sulfate.
at translational or post-translation levels and is implicated in synaptic development and plasticity. 1998 IBRO. Published by Elsevier Science Ltd. We prepared primary cultures of neocortical neurons from fetal rats and grew them in serum-free medium with daily applications of neurotrophins (50 ng/ml). The serum-free culture condition prevented the proliferation of non-neuronal cells so that they accounted for less than 5% of the total number of cells.27 Applications of brain-derived neurotrophic factor (BDNF) markedly increased the protein expression of GluR1 and GluR2/3 receptors. By setting the average control receptor level as 100 for each immunoblot, their protein levels were found to change from 10012% to 19322% for GluR1, and from 1006% to 18211% for GluR2/3 (n=3; P<0.05). A typical immunoblot is shown in Fig. 1. In contrast, the other neurotrophins, nerve growth factor (NGF) and neurotrophin-3, failed to alter the protein levels of the AMPA receptors. Moreover, expressions of the N-methyl--aspartate (NMDA) receptor 1 and 2A/B subunits were not significantly affected: control levels and the BDNF-stimulated levels were 1005% and 1019% for NMDA receptor-1 (NMDAR1) protein, and 1008% and 1205% for NMDAR2A/B protein, respectively. In addition, other cytoskeletal proteins, á-tubulin and neurofilament protein, were not affected. Importantly, K252b (1 µM), an inhibitor of receptor tyrosine kinases, blocked the effects of BDNF. Thus, these results suggest that the BDNF activity is quite specific for the AMPA receptors. To analyse the effect of BDNF on individual neurons, neocortical cultures were immunostained with anti-GluR1 and anti-GluR2/3 antibodies (Fig. 2). Although the variation in staining intensity was substantial, about half of the neurons in control
1009
1010
M. Narisawa-Saito et al.
Fig. 1. Effects of neurotrophins on glutamate receptor proteins in cultured rat neocortical neurons. Proteins were extracted from embryonic neocortical neurons grown for five days either with or without () daily applications of 50 ng/ml NT-3, BDNF and NGF, and BDNF plus 1 µM K252b. The protein was separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, transferred to a membrane, and probed with antibodies directed against GluR1, GluR2/3, NMDAR1, NMDAR2A/B (all recognize their carboxyl termini; Chemicon), á-tubulin (SANBIO) and anti-neurofilaments (NF) 200/150,000 mol. wt (a gift from Dr Takahashi, Niigata University, Japan). Immunoreactivity was visualized by chemiluminescence (ECL kit; Amersham). Whole cerebral neocortices of embryonic day-18 rats were mechanically dissociated and plated onto laminin-coated dishes. Cortical neurons were grown with Dulbecco’s modified Eagle’s medium containing 0.5 mM purified glutamine (Ajinomoto Co), nutrient mixture N2 and 10 mM HEPES (pH 7.3).27 The mouse 2.5S NGF was obtained from Boehringer Mannheim, and BDNF and NT-3 were from Amgen (Thousand Oaks, CA, U.S.A.).
culture exhibited a moderate level of GluR1- or GluR2/3-immunoreactivity (Table 1), which is almost consistent with previous in vivo immunohistochemical observations.8 Cells with strong immunoreactivity in their perikarya were only rarely seen. After growth
with BDNF, however, the frequency of neocortical neurons that contained strong immunoreactivity for either GluR1 or GluR2/3 more markedly increased than that exhibiting the moderate immunoreactivity for the receptors. Total cell densities were not influenced by BDNF, suggesting that BDNF did not influence neuronal survival under the culture conditions used.27 To test whether BDNF enhances the mRNA expression of the AMPA receptors, we measured their GluR mRNA levels by RNA blotting (Fig. 3). Interestingly, the AMPA receptor mRNA levels were not significantly altered by treatment with BDNF: mRNA changes over control cultures were calculated as 995% for GluR1 mRNA, 955% for GluR2 mRNA and 9910% GluR3 mRNA when cyclophilin mRNA was used as an internal control (n=3). We also carried out the RNase protection assay and confirmed the equality in GluR1 mRNA: GluR1 mRNA levels were 93.45.9% in the BDNF-treated culture in comparison with 1002.5% in control culture (n=4). Again, these results suggest that the regulation of AMPA receptor expression mainly involves modulating translational and/or post-translational mechanisms. Analyses of BDNF-knockout mice have confirmed the idea that BDNF is essential for normal neural development12,18 and synaptic transmission.21,28 We also examined BDNF-knockout mice to study the effect of BDNF on AMPA receptor proteins in vivo. The neocortical GluR levels of the BDNF knockout mice exhibited decreasing tendency compared to those of wild mice but the difference did not reach statistical significance in total protein fractions (data not shown). Thus, we fractionated the subcellular components of the neocortical tissue into membrane (S), postsynaptic density/cytoskeletal (P) and cytoplasmic fractions according to the method of McNeill and Colbran26 (Fig. 4A). More than 90% of GluR2/3 immunoreactivity was recovered in the S or P fraction. Between the S and P fractions, immunoreactivity for á-calmodulin-dependent kinase II (áCAMKII), a marker for postsynaptic density (PSD),20 was fractionated into the P fraction with high efficiency (Table 2). The BDNF-knockout mice exhibited a marked reduction in GluR2/3 immunoreactivity (P<0.05) in the P fraction, but not in the S fraction (Fig. 4B, Table 2). In contrast, levels of áCAMKII in the P fractions of both genotypes were almost same. These observations suggest that endogenous BDNF specifically influences the subcellular expression of the AMPA receptors in vivo. DISCUSSION
In the present study, our results suggest that BDNF can specifically increase protein levels of AMPA receptors. Reversibly, expression of BDNF itself is known to be induced by activation of AMPA
AMPA receptor phenotype regulated by BDNF
1011
Fig. 2. Immunocytochemical analysis of the BDNF effects on AMPA receptors. Cultures of rat neocortical neurons were prepared similarly and grown for five days with or without BDNF. Paraformaldehyde-fixed cells were processed for immunostaining with anti-AMPA receptor (GluR1 and GluR2/3) antibodies as well as with anti-NMDA receptor (NMDAR1 and NMDAR2A/B) antibodies. Antigen–antibody complexes were visualized by the ABC method, with nickel enhancement.27 Scale bar=100 µm.
receptors in the brain.11,35 Therefore, these two observations indicate that BDNF and AMPA receptors mutually interact with each other to regulate positively the other expression. Their interactions might contribute to normal brain development and plasticity.23,33 The receptor increase appeared to occur within only a part of neuronal populations because some neurons showed no response to BDNF. We have previously described a similar phenotypic regulation of neocortical neurons by neurotrophins: BDNF upregulates peptidergic expression in neocortical and striatal neurons in vivo and in vitro without changing their survival rate.10,27 All these phenotypes are associated with nonpyramidal GABAergic neurons
and are impaired in the neocortex of BDNFknockout mice.18 Interestingly, most of the neocortical neurons that show strong GluR1 immunoreactivity are also GABAergic nonpyramidal neurons,8 some of which express higher levels of BDNF receptor (trkB) mRNA as well.6 These results suggest that BDNF may act mainly on this GABAergic neuronal population to result in an increase in the AMPA receptor proteins. Our RNA analyses indicated that the actions of BDNF on the receptor expression presumably involve translational or post-translational processes. This BDNF activity appears to be different from the reported activities of other neurotrophic factors, such as Neu differentiation factor which induces mRNA
1012
M. Narisawa-Saito et al. Table 1. Increased frequency of AMPA-receptor immunoreactive neurons in culture GluR1 Positive
GluR2/3 Positive
Treatment
Moderate
Strong
Moderate
Strong
Cell density (cells/mm2)
None BDNF
47.82.9 43.72.3
4.20.5 33.53.2**
46.31.4 59.22.1**
6.20.8 12.51.9*
75744 85645
Six sister cultures of neocortical neurons were prepared as described in Fig. 2. Since the intensity of positive immunoreactivity was heterogenous among neurons, immunoreactive cells were classified into two groups (cells with moderate staining and those with strong staining) and their numbers were counted. The frequency of these positive cells (%) was calculated and presented. Statistical analysis was performed with a one-way analysis of variance (ANOVA). Results are expressed as meansS.E.M. (n=6). *P<0.05, **P<0.01 compared with control. Note: cell densities were not significantly altered by BDNF.
Fig. 3. BDNF effects on mRNA levels for the AMPA receptors. RNA was separated using a 1% agarose gel, transferred to a membrane, and probed with 32P-labeled cDNA for GluR1, GluR2 and GluR3.17 The RNA extracted from rat whole brains was used as a positive control. A typical example of a RNA blot is shown for display. 32P-labeled cDNA probes for rat GluR1 (a 2.2-kb fragment of p59/2; a gift from Dr Heinemann, The Salk Institute, CA, U.S.A.), mouse GluR2 (a 1.5-kb fragment; a gift from Dr Sakimura, Niigata University, Japan), rat GluR3 (a 269nt–1337nt fragment produced by reverse transcriptionpolymerase chain reaction) and rat cyclophilin (a 0.7-kb fragment) were generated using the Random Primed DNA Labeling kit (Boehringer). These probes were sequentially hybridized to three RNA filters (n=3) and the relative signal intensity of GluR mRNAs to that of cyclophilin mRNA was determined. Alternatively, an RNase protection assay was performed by using the RNase protection kit (Boehringer Mannheim) with a 367-base riboprobe that was made from T7 promoter to a BalI site of the rat GluR1 plasmid (data not shown).
for neurotransmitter receptors.30 Post-translational modifications of transmitter receptors at the neuromuscular synapse have been extensively studied. The non-receptor-type tyrosine kinase, Src, and musclespecific kinase are associated with nicotinic acetylcholine receptors, and their activation has been implicated in receptor clustering and stabilization at postsynaptic sites by altering their cytoskeletal inter-
actions.14,15,32 The BDNF receptor trkB and AMPA receptors are all intrinsic components of PSD.9,29,31 Thus, it is quite natural to speculate that the protein phosphorylation cascade, starting with the activation of the trkB kinase, may contribute to the regulation of AMPA receptor stability in the brain. However, we cannot deny a possibility that BDNF might also affect gene transcription of the receptors whose
AMPA receptor phenotype regulated by BDNF
1013
Table 2. Subcellular fractionation of synaptic proteins from the neocortex of wild-type and brain-derived neurotrophic factor-knockout mice. Protein yield Molecule áCAMKII SNAP25 GluR2/3
Fig. 4. The expression of GluR2/3 protein in the subcellular fractions from the neocortex of BDNF-knockout mice. (A) The subcellular fractionation carried out using the method of McNeill and Colbran.26 The neocortical homogenates were separated by centrifugation (100,000 g for 60 min) into the supernatant (cytosolic proteins) and the pellet containing crude membranes. The pellet was rehomogenized with 1% Triton X-100-containing buffer and centrifuged as before to give rise to supernatant 2 (S; membrane fraction) and precipitate 2 (P; PSD/cytoskeletal fraction). Authenticity of the fractionation was examined by western blotting of the protein markers, SNAP257 (a presynaptic membrane protein) and áCAMKII20 (a PSD protein) as shown in Table 2. Each lane contained 20 µg protein of the fraction. (B) BDNF-knockout mice12 were produced by crossing heterozygous mutant mice and grown until postnatal day 12–14. The subcellular proteins of each mouse neocortex were fractionated by the above method into S and P fractions, and protein levels of GluR2/3 in these fractions were similarly examined by western blotting. A window containing each pair of fractions represents the data from a wild mouse or a homozygous mouse (-/-) (n=3 animals for each genotype).
mRNA changes would be transient and disappear acutely. The neurotrophins can modulate basal synaptic transmission18,34 as well as long-term potentia-
Fraction
Wild (%)
S P S P
22.44.7 77.63.2** 86.510.4** 13.51.5
S P
5.00.9 1006.1
Homozygotes (%) 70.75.7
5.30.4 51.711.5*
Authenticity of the subcellular fractionation was ascertained by measuring the recoverey ratio of the marker proteins, SNAP25[7] and áCAMKII[20] (n=3) as shown in Fig. 4A. In addition, measuring the intensity of signals on western blots in Fig. 4B, levels of GluR2/3 protein as well as those of áCAMKII in the neocortex were compared between wild and BDNF-knockout mice (n=3). In the comparison for GluR2/3, the highest protein level was set as 100%. Statistical analysis was performed with a one-way ANOVA followed by the Bonferroni test. Results are expressed as meansS.E.M. **P<0.01 compared between the S and P fractions, and *P<0.05 compared between wild-type and the mutant mice. Note: protein levels of áCAMKII were similar in the P fraction of both genotypes.
tion.1,13 Changes in hippocampal neurotransmission induced by BDNF similarly depend upon translation, but not on transcription.19 Subsequent electrophysiological studies will illuminate how the total increase in the AMPA receptor expression is correlated with the acute BDNF-mediated changes of neurotransmission that were reported previously. In conclusion, the present results will provide novel insights into the mechanisms involved in the regulation of AMPA receptor expression during development and in synaptic plasticity. Acknowledgements—We thank Dr P. Ernfors and Dr R. Jaenisch for providing the brain tissues from BDNF mutant mice, Miss J. H. Kogan and Mrs C. Fujikawa for technical assistance, Dr S. Heinemann for rat GluR1 cDNA, Dr K. Sakimura for mouse GluR2 cDNA and GluR2/3 antibodies, and Dr T. Yamamoto for his advice. This project was supported by the Japan Society for the Promotion of Science (RFTF-96L00203) and Uehara Memorial Foundation.
REFERENCES
1.
Akaneya Y., Tsumoto T. and Hatanaka H. (1996) Brain-derived neurotrophic factor blocks long-term depression in rat visual cortex. J. Neurophysiol. 76, 4198–4201. 2. Barde Y. A., Edgar D. and Thoenen H. (1982) Purification of a new neurotrophic factor from mammalian brain. Eur. molec. Biol. Org. J. 1, 549–553. 3. Berkemeier L. R., Winslow J. W., Kaplan D. R., Nikolics K., Goeddel D. V. and Rosenthal A. (1991) Neurotrophin-5: a novel neurotrophic factor that activates trk and trkB. Neuron 7, 857–866. 4. Cabelli R. J., Shelton D. L., Segal R. A. and Shatz C. J. (1997) Blockade of endogenous ligands of trkB inhibits formation of ocular dominance columns. Neuron 19, 63–76. 5. Causing C. G., Gloster A., Aloyz R., Bamji S. X., Chang E., Fawcett J., Kuchel G. and Miller F. D. (1997) Synaptic innervation density is regulated by neuron-derived BDNF. Neuron 18, 257–267.
1014 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
M. Narisawa-Saito et al. Cellerino A., Maffei L. and Domenici L. (1996) The distribution of brain-derived neurotrophic factor and its receptor trkB in parvalbumin-containing neurons of the rat visual cortex. Eur. J. Neurosci. 8, 1190–1197. Chapman E. R., An S., Barton N. and Jahn R. (1994) SNAP-25, a t-SNARE which binds to both syntaxin and synaptobrevin via domains that may form coiled coils. J. biol. Chem. 269, 27,427–27,432. Conti F., Minell A. and Brecha N. C. (1994) Cellular localization and laminar distribution of AMPA glutamate receptor subunits mRNAs and proteins in the rat cerebral cortex. J. comp. Neurol. 350, 241–259. Craig A. M., Blackstone C. D., Huganir R. L. and Banker G. (1993) The distribution of glutamate receptors in cultured rat hippocampal neurons: postsynaptic clustering of AMPA-selective subunits. Neuron 10, 1055–1068. Croll S. D., Wiegand S. J., Anderson K. D., Lindsay R. M. and Nawa H. (1994) Regulation of neuropeptides in adult rat forebrain by the neurotrophins BDNF and NGF. Eur. J. Neurosci. 6, 1343–1353. Ernfors P., Bengzon J., Kokaia Z., Persson H. and Lindvall O. (1991) Increased levels of messenger RNAs for neurotrophic factors in the brain during kindling epileptogenesis. Neuron 7, 165–176. Ernfors P., Lee K. F. and Jaenisch R. (1994) Mice lacking brain-derived neurotrophic factor develop with sensory deficits. Nature 368, 147–150. Figurov A., Pozzo-Miller L. D., Olafsson P., Wang T. and Lu B. (1996) Regulation of synaptic responses to high-frequency stimulation and LTP by neurotrophins in the hippocampus. Nature 381, 706–709. Fuhrer C. and Hall Z. W. (1996) Functional interaction of Src family kinases with the acetylcholine receptor in C2 myotubes. J. biol. Chem. 271, 32,474–32,481. Fuhrer C., Sugiyama J. E., Taylor R. G. and Hall Z. W. (1997) Association of muscle-specific kinase MuSK with the acetylcholine receptor in mammalian muscle. Eur. molec. Biol. Org. J. 16, 4951–4960. Hohn A., Leibrock J., Bailey K. and Barde Y. A. (1990) Identification and characterization of a novel member of the nerve growth factor/brain-derived neurotrophic factor family. Nature 344, 339–341. Hollmann M. and Heinemann S. (1994) Cloned glutamate receptors. A. Rev. Neurosci. 17, 31–108. Jones K. R., Farinas I., Backus C. and Reichardt L. F. (1994) Targeted disruption of the BDNF gene perturbs brain and sensory neuron development but not motor neuron development. Cell 76, 989–999. Kang H. and Schuman E. M. (1996) A requirement for local protein synthesis in neurotrophin-induced hippocampal synaptic plasticity. Science 273, 1402–1406. Kennedy M. B. (1997) The postsynaptic density at glutamatergic synapses. Trends Neurosci. 20, 264–268. Korte M., Griesbeck O., Gravel C., Carroll P., Staiger V., Thoenen H. and Bonhoeffer T. (1996) Virus-mediated gene transfer into hippocampal CA1 region restores long-term potentiation in brain-derived neurotrophic factor mutant mice. Proc. natn. Acad. Sci. U.S.A. 93, 12,547–12,552. Levi-Montalcini R. (1987) The nerve growth factor: thirty-five years later. Eur. molec. Biol. Org. J. 6, 1145–1154. Lewin G. R. and Barde Y. A. (1996) Physiology of the neurotrophins. A. Rev. Neurosci. 19, 289–317. Maisonpierre P. C., Belluscio L., Squinto S., Ip N. Y., Furth M. E., Lindsay R. M. and Yancopoulos G. D. (1990) Neurotrophin-3: a neurotrophic factor related to NGF and BDNF. Science 247, 1446–1451. McAllister A. K., Katz L. C. and Lo D. C. (1997) Opposing roles for endogenous BDNF and NT-3 in regulating cortical dendritic growth. Neuron 18, 767–778. McNeill R. B. and Colbran R. J. (1995) Interaction of autophosphorylated Ca2+/calmodulin-dependent protein kinase II with neuronal cytoskeletal proteins. J. biol. Chem. 270, 10,043–10,049. Nawa H., Bessho Y., Carnahan J., Nakanishi S. and Mizuno K. (1993) Regulation of neuropeptide expression in cultured cerebral cortical neurons by brain-derived neurotrophic factor. J. Neurochem. 60, 772–775. Patterson S. L., Abel T., Deuel T. A. S., Martin K. C., Rose K. C. and Kandel E. C. (1996) Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron 16, 1137–1145. Petralia R. S. and Wenthold R. J. (1992) Light and electron immunocytochemical localization of AMPA-selective glutamate receptors in the rat brain. J. comp. Neurol. 318, 329–354. Sandrock A. W. Jr, Dryer S. E., Rosen K. M., Gozani S. N., Kramer R., Theill L. E. and Fischbach G. D. (1997) Maintenance of acetylcholine receptor number by neuregulins at the neuromuscular junction in vivo. Science 276, 599–603. Suen P. C., Wu K., Levine E. S., Mount H. T., Xu J. L., Lin S. Y. and Black I. B. (1997) Brain-derived neurotrophic factor rapidly enhances phosphorylation of the postsynaptic N-methyl--aspartate receptor subunit 1. Proc. natn. Acad. Sci. U.S.A. 94, 8191–8195. Swope S. L. and Huganir R. L. (1993) Molecular cloning of two abundant protein tyrosine kinases in Torpedo electric organ that associate with the acetylcholine receptor. J. biol. Chem. 268, 25,152–25,161. Thoenen H. (1995) Neurotrophins and neuronal plasticity. Science 270, 593–598. Tanaka T., Saito H. and Matsuki N. (1997) Inhibition of GABA-A synaptic responses by brain-derived neurotrophic factor (BDNF) in rat hippocampus. J. Neurosci. 17, 2959–2966. Zafra F., Hengerer B., Leibrock J., Thoenen H. and Lindholm D. (1990) Activity dependent regulation of BDNF and NGF mRNAs in the rat hippocampus is mediated by non-NMDA glutamate receptors. Eur. molec Biol. Org. J. 9, 3545–3550. (Accepted 21 August 1998)