- Email: [email protected]
Differential regulation of MAP2 and αCamKII expression in hippocampal neurones by forskolin and calcium ionophore treatment
Molecular Brain Research 122 (2004) 10 – 16 www.elsevier.com/locate/molbrainres
Research report
Differential regulation of MAP2 and aCamKII expression in hippocampal neurones by forskolin and calcium ionophore treatment K.A. Alier, B.J. Morris * Division of Neuroscience and Biomedical Systems, Institute of Biomedical and Life Sciences, University of Glasgow, West Medical Building, Glasgow G12 8QQ, UK Accepted 19 November 2003
Abstract The genes encoding microtubule-associated protein 2 (MAP2), and the alpha subunit of calcium/calmodulin-dependent protein kinase II (aCaMKII), are members of a small number of genes whose expression is increased in hippocampal neurones during the intermediate phase of long-term potentiation (LTP)—a phase dependent on mRNA translation but not on gene transcription. However, the intracellular signalling pathways which mediate these increases in expression are largely unknown. Organotypic slice cultures of rat hippocampus were exposed to either forskolin (to elevate cAMP levels), A23187 (to increase intracellular Ca2 + levels) or the corresponding vehicle. The levels of immunoreactive (ir-) MAP2 were increased 4 h after forskolin treatment, but were unaffected by A23187 treatment. Conversely, the levels of ir-aCaMKII were increased 4 h after A23187 treatment, but were unaffected by forskolin. The regulation of the expression of these proteins was the same in the CA3 region as in the CA1 and dentate gyrus of the hippocampus. While rapamycin reduced the basal levels of ir-MAP2, it did not affect the ability of either forskolin or A23187 to enhance ir-MAP2 or ir-aCaMKII levels. These results suggest that cAMP and Ca2 + differentially modulate the expression of these two plasticity-related genes, and that translational enhancement via the mammalian target of rapamycin kinase is not involved in these effects. D 2004 Elsevier B.V. All rights reserved. Keywords: Gene expression; Hippocampus; Microtubule-associated protein; Calcium/calmodulin-dependent protein kinase II; cAMP; Rapamycin
1. Introduction Long-term potentiation (LTP) of synaptic transmission is observed after high frequency stimulation of pathways in the hippocampal formation. In the dentate gyrus (DG) and CA1 regions, the induction of LTP is dependent on activation of the NMDA class of glutamate receptor, while in the CA3 region, it is primarily dependent on increased cAMP levels [2]. LTP, in common with most other forms of synaptic plasticity, can be divided into three temporal phases. The earliest phase is sustained by modification of existing proteins, for example, by phosphorylation or subcellular redistribution. The latest phase is sustained by de novo gene transcription and mRNA translation. The intermediate phase, which appears between 1 and 6 h after the initial stimulation, depending on the experimental
* Corresponding author. Tel.: +44-141-330-5361; fax: +44-141-3305659. E-mail address: [email protected] (B.J. Morris). 0169-328X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.molbrainres.2003.11.018
model, is dependent on mRNA translation and de novo protein synthesis, but not on gene transcription [15,16]. There is hence considerable interest in identifying the proteins which are synthesised at this time in response to the stimulation, and the mechanisms which regulate their expression. A handful of mRNAs are present in neuronal dendrites, in the vicinity of the synapses, in addition to the cell soma. Two of these mRNA species, encoding MAP2 and aCaMKII, have been associated with various forms of neuronal plasticity [7– 12,17 –20]. The role of MAP2 in cross-linking microtubules and in facilitating changes in dendritic architecture [3,4,6] makes it an attractive candidate for enabling the synaptic remodelling associated with LTP [5], while CaMKII plays a central role in transducing the effects of NMDA receptor stimulation into lasting intracellular effects [13,17,20]. Increased levels of MAP2 mRNA and aCaMKII mRNA were detected after LTP induction [11,12,14,18,19,24], at least in the dentate gyrus and CA1 regions of the hippocampus. It is not yet known if similar increases are observed
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
in the CA3 region. Current evidence suggests that the increases arise from post-transcriptional regulation of existing mRNA in the dendrites, local to the region of stimulation [16,23]. This local increase in mRNA levels is then translated into elevated levels of the corresponding protein. However, it is unclear what mechanisms couple glutamatergic activation of NMDA receptors, which trigger the induction of LTP, to enhanced MAP2 and aCaMKII mRNA levels. Two major effectors of NMDA receptor activation are Ca2 +, which enters the cell via the activated receptor, and cAMP, which is elevated by the activation of adenyl cyclase. In this paper, we have used hippocampal neurones, maintained in organotypic culture, to study the effect on the levels of MAP2 and aCaMKII of agents which directly increase Ca2 + or cAMP levels. We have tested the hypothesis that Ca2 + influx will be the primary stimulus increasing the expression of both the MAP2 and aCaMKII genes, and that the effects observed in the dentate gyrus and CA1 regions may not be mirrored in the CA3 region.
11
2. Materials and methods 2.1. Cell culture Rat neonates between the ages of 6 and 10 days old were killed using 0.2 ml/animal of an anaesthetic (Euthatal, Rhone Merieux). The hippocampus was removed rapidly and 250 Am hippocampal slices were obtained using a tissue chopper (McIlwain). The slices were then placed in six-well plates (Iwaki) with porous (0.2 Am), transparent, and low protein binding inserts (Falcon) as described [22]. The wells contained 1 ml of Wilde medium each [Basal Medium Eagle (BME) 1 liquid with Earle’s salt without L-glutamine (88.5%), horse serum heat inactivated (17.7%), glucose (22.5 g/l), glutamaxk-1 Supplement (0.46%), penicillin/streptomycin solution (1:106.5 dilutions)]. All the contents of the Wilde medium were purchased from Life Technologies. The six-well plates were then placed in an incubator (95% O2 and 5% CO2) for 4 –5 days.
Fig. 1. The effect of A23187 and forskolin on ir-aCamKII and ir-MAP2 in hippocampal slice cultures. Upper panels: Photomicrographs showing detection of ir-aCamKII after treatment with vehicle (a, b, c) or A23187 (d, e, f), in the dentate gyrus (a, d), CA1 (b, e) and CA3 (c, f) regions. Arrow heads and arrows indicate dendrite and somata, respectively. Note the enhanced staining in all regions after A23187 treatment. Lower panels: Photomicrographs showing detection of ir-MAP2 after treatment with vehicle (a, b, c) or forskolin (d, e, f), in the dentate gyrus (a, d), CA1 (b, e) and CA3 (c, f) regions with the arrow heads and arrows representing the dendrites and somata, respectively. Note the enhanced staining in all regions after forskolin treatment. Scale bar: 20 Am for CA1 and CA3, 15 Am for DG.
12
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
2.2. Hippocampal slice treatments In a first series of experiments, the slices were treated with vehicle [Hanks Balance Salt Solution (HBSS), Life Technologies], forskolin (Sigma, 50 AM) or A23187 (Sigma, 5 AM) in vitro. In a second series of experiments, the slices were treated with vehicle or rapamycin (Calbiochem, 20 nM) for 30 min on the fifth day in vitro before treatment with vehicle, forskolin or A23187. 2.3. Immunocytochemistry Four hours after addition of vehicle, forskolin or A23187, the treated slices were fixed in 3% formaldehyde (Riedel-de Haen) before processing for immunocytochemistry as previously described [16] with monoclonal antibodies specific for aCaMKII (Boehringer, 1:1000) and MAP2 (Sigma, 1:1000) using standard methods. Detection of primary antibody was achieved using biotinylated secondary antibody (anti-mouse) and the ABC (Vector Laboratories) with VIP (Vector Laboratories). The stain-
ing was quantified by using ‘‘Image’’ software (W. Rasband, NIH) as described [21]. The intensity of staining in drug-treated slices was expressed as a percentage of the staining in parallel vehicle-treated slices. The background staining was subtracted to obtain a specific staining using 10 random fields of measurement of different cell bodies/animal. The significance between groups was determined using analysis of variance with post hoc Fisher’s test for multiple pairwise comparison using Minitab statistical program. Significance relative to 100% was assessed by 95% confidence intervals of the mean. 2.4. Western blot Four hours after addition of vehicle, forskolin or A23187, cultured slices were scraped from the inserts and homogenised in RIPA buffer [1 M Tris, pH 8, 1 M NaCl, HP40, 0.5% DOC, 10% SDS, protease inhibitors (Roche Diagnostics)] and centrifuged at 10,000 rpm. After the second centrifugation, the supernatant was denatured and run in NuPAGEk 12% Bis – Tris gels with rainbow
Fig. 2. Semi-quantitative analysis of the effect of A23187 and forskolin on ir-aCamKII in cell bodies (a) or dendrites (b), and ir-MAP2 in cell bodies (c) or dendrites (d) in hippocampal slice cultures. Results are shown for n = 15, *P < 0.05 vs. vehicle treatment ( = 100%) by 95% confidence intervals.
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
13
markers (Amersham Life Science). After blotting into PVDF membrane (Invitrogen), the proteins were visualised with primary antibodies (same as above), HRP secondary antibody (Diagnostics Scotland) and ECL detection kit (Amersham Life Sciences) according to standard procedures. Once the required bands were obtained, the blot was stripped using a stripping buffer [80 mM Tris, pH 6.8, 2% SDS and 68% h-mercaptoethanol (purchased from Sigma)] and reprobed with h-actin (1:5000, Sigma) primary antibody.
3. Results 3.1. The effect of A23187 and forskolin on the expression of ir-aCaMKII and ir-MAP2 Neurones in all hippocampal subregions were stained with the aCaMKII and MAP2 antibodies, with staining prominent in both cell body and dendritic layers. The Ca2 + ionophore, A23187, increased the levels of iraCaMKII in DG, CA3 and CA1 regions of the slice cultures by between 15 –25% in the somata and 9 – 18% in apical dendrites. However, the adenylate cyclase activator, forskolin, did not have any significant effect on iraCaMKII levels in any region (Figs. 1 and 2a,b). Conversely, levels of ir-MAP2 expression were elevated in both cell bodies and dendrites all three regions of the hippocampus in the presence of forskolin while A23187 did not have any significant effect (Figs. 1 and 2c,d). Western blot analysis also indicated an increase in the expression of CaMKII in the presence of A23187 while MAP2 levels were increased in the presence of forskolin (Fig. 3).
Fig. 4. The differential effect of rapamycin on aCaMKII levels in vehicle-, A23187- and forskolin-treated slices: ir-aCaMKII levels were monitored in the cell bodies (a) and dendritic regions (b). Results are expressed as a percentage of the immunocytochemical signal in the corresponding region after vehicle/vehicle treament. ir-aCaMKII levels were not affected by rapamycin pretreatment, but were significantly increased in all three regions of the hippocampus by A23187 (*P < 0.05 versus corresponding region after vehicle treatment in the absence of rapamycin, 95% confidence intervals of the mean). The expression of ir-aCaMKII in DG, CA3 and CA1 was significantly increased in rapamycin/A23187treated slices compared to rapamycin/vehicle ( + P < 0.05, ANOVA with post hoc Fisher test).
3.2. Effect of rapamycin on the expression of ir-aCaMKII and ir-MAP2 in both stimulated and unstimulated hippocampal slices
Fig. 3. The effects of A23187 and forskolin on the hippocampal iraCaMKII and ir-MAP2 were further examined by Western analysis. Bands of predicted sizes were identified (280 and 70 kDa for MAP2, 50 kDa for aCaMKII). The blots were stripped and reprobed with 42 kDa h-actin which was used as an internal standard. The bands for h-actin were uniform, indicating an equal concentration of protein.
The levels of ir-aCaMKII in cell bodies and dendrites were not significantly affected by pretreatment of slices with rapamycin in any of the hippocampal regions (Fig. 4a,b). For A23187-treated hippocampal slices, there was a significant increase in the levels of ir-aCaMKII in both vehicletreated and rapamycin-pretreated slices. This increase occurred in DG, CA1 and CA3 regions (Fig. 4a,b). This indicated that rapamycin had no effect on the elevation of ir-aCaMKII induced by A23187. There was a significant decrease in the basal levels irMAP2 after pretreatment of slices with rapamycin in the
14
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
Fig. 5. The differential effect of rapamycin on MAP2 levels in vehicle-, A23187- and forskolin-treated slices: ir-MAP2 levels were monitored in the cell bodies (a) and dendritic regions (b). Results are expressed as a percentage of the immunocytochemical signal in the corresponding region after vehicle/vehicle treament. The basal expression of ir-MAP2 was significantly reduced by rapamycin in all three regions of the hippocampus (*P < 0.05 versus vehicle treatment without rapamycin, 95% confidence intervals of the mean), ir-MAP2 levels were significantly increased in all three regions of the hippocampus by forskolin (*P < 0.05 versus corresponding region after vehicle treatment in the absence of rapamycin, 95% confidence intervals of the mean). The expression of ir-MAP2 in DG, CA3 and CA1 was significantly increased in rapamycin/forskolin-treated slices compared to rapamycin/vehicle ( f P < 0.05, ANOVA with post hoc Fisher test).
somata (Fig. 5a) compared to slices receiving vehicle pretreatment. Forskolin treatment increased ir-MAP2 levels (in somata and dendrites) relative to vehicle treatment, even after pretreatment with rapamycin, indicating that rapamycin had no effect on the elevation of ir-MAP2 in all three regions of the hippocampus during stimulation (Fig. 5).
4. Discussion The results indicate that treatment with A23187 increases hippocampal ir-aCaMKII while forskolin has
no effect. This suggests the importance of Ca2 + influx in regulating the expression of aCaMKII. One possible mechanism by which Ca2 + influx stimulates aCaMKII synthesis is via phosphorylation of cytoplasmic polyadenylation element binding protein (CPEB), which is present in the hippocampal dendrites and somata [25]. Wu et al. [25] provided evidence that binding of CPEB to CPE sites located in the 3V-end of the RNA message for aCaMKII can stimulate its translation rate. However, as yet there is no evidence that CPEB is phosphorylated in response to Ca2 + influx. The Ca2 + ionophore (A23187) does not produce any significant effect on ir-MAP2 levels. However, elevation of ir-MAP2 levels is observed after forskolin treatment, suggesting the involvement of cAMP in the regulation of MAP2 expression. Forskolin activates adenylate cyclase (AC), which increases the levels of cAMP. AC2 and AC4 have been shown to be expressed in the hippocampal formation and are co-localised with MAP2 [1]. AC2 and AC4 have been labelled in the dendrites and cell bodies of DG, CA3 and CA1 [1]. Hence these two isoforms are likely to be involved in the expression of ir MAP2. Therefore cAMP and PKA might act on the MAP2 mRNA via an unknown mechanism resulting in the elevation of MAP2 in the three regions of the hippocampus. These results provide strong evidence that the elevation of ir aCaMKII and MAP2 associated with synaptic plasticity may be achieved via separate pathways. This in turn would suggest that the two genes are not necessarily regulated in parallel, but could show a differential relative change in expression according to the relative change in Ca2 + and cAMP levels. The data also suggests that the elevation of mRNA in the dendrites plays a role in the expression of these proteins in the dendritic region of the hippocampus. The levels of the two proteins were elevated in all the three regions of the hippocampus i.e. DG, CA3 and CA1, suggesting that the intracellular pathways regulating MAP2 and aCaMKII expression in CA3 are similar to those in CA1 and DG. This is the first evidence that CA3 neurones can show enhanced MAP2 and CaMKII expression in the same way as DG and CA1 neurones. Rapamycin produces its effect by binding to and inhibiting a kinase, the mammalian target of rapamycin (mTOR). It is well known that mTOR can activate mRNA translation via phosphorylation of ribosomal accessory proteins. Rapamycin hence acts to switch off translation resulting in protein synthesis inhibition. Rapamycin reportedly inhibits the intermediate phase of LTP [23], and so mTOR could be involved in stimulating translation of proteins such as MAP2 and aCaMKII. Rapamycin did not have any significant effect on either the basal expression of aCaMKII (Fig. 4) or the increased expression after A23187 treatment. Hence the regulation of aCaMKII expression appears to be achieved via a rapamycin-insensitive pathway. However, rapamycin decreased the basal expression of ir-MAP2 significantly, but did not have any effect on ir-MAP2 stimulation by forskolin (Fig. 5),
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
suggesting that at basal levels inhibition of MAP2 expression is achieved via a rapamycin-sensitive pathway while after stimulation the change in expression occurs through a rapamycin-insensitive pathway. This suggests that aCaMKII CaMKII and MAP2 are not among any rapamycin-sensitive genes involved in the intermediate phase of LTP, and is consistent with the concept that post-translational activation of these genes occurs through other pathways, such as mRNA stabilisation [16]. Our initial hypothesis—that Ca2 + influx would be the primary stimulus increasing the expression of both the MAP2 and aCaMKII genes, and that the effects observed in the dentate gyrus and CA1 regions might not be mirrored in the CA3 region—therefore proved over-simplistic. Clearly Ca2 + influx is not the primary stimulus regulating all plasticity-related genes, and cAMP can play a major role distinct from Ca2 + influx. Furthermore, we found the regulation of these genes to be identical in the CA3 region as compared to the dentate gyrus and CA1 region. Hence while the mechanisms for triggering the plasticity response are distinct in the CA3 region, the mechanisms regulating the expression of late-response genes appear to be the same. This implies that there may be many commonalities in the genomic changes in different models of late-phase synaptic plasticity.
5. Conclusion Ca2 + plays an important role in the expression of iraCaMKII while an increase in the levels of cAMP affects the levels of expression of ir-MAP2. Intracellular pathways for CA3 are similar to DG and CA1 since the levels of both proteins were elevated in all the three regions of the hippocampus. Basal levels of ir-MAP2 follow the rapamycin-sensitive pathway while the stimulated levels of ir-MAP2 and both basal and stimulated levels ir-aCaMKII follow the rapamycin-insensitive pathway, suggesting different post-translational activation pathways.
Acknowledgements This work would not have been possible without the support of the Africa Educational Trust (AET)—we are very grateful for their support.
References [1] L.P. Baker, M.D. Nielsen, S. Impey, B.M. Hacker, S.W. Poser, M.Y. Chan, D.R. Storm, Regulation and immunohistochemical localization of betagamma-stimulated adenylyl cyclases in mouse hippocampus, J. Neurosci. 19 (1999) 180 – 192. [2] T.V.P. Bliss, G.L. Collingridge, A synaptic model of memory: longterm potentiation in the hippocampus, Nature 361 (1993) 31 – 38.
15
[3] A. Caceres, J. Busciglio, A. Ferreira, O. Steward, An immunocytochemical and biochemical study of the microtubule-associated protein MAP-2 during post-lesion dendritic remodeling in the central nervous system of adult rats, Brain Res. 427 (1988) 233 – 246. [4] A. Caceres, J. Mautino, K.S. Kosik, Suppression of MAP2 in cultured cerebellar macroneurons inhibits minor neurite formation, Neuron 9 (1992) 607 – 618. [5] F.L. Chang, W.T. Greenough, Transient and enduring morphological correlates of synaptic activity and efficacy change in the rat hippocampal slice, Brain Res. 309 (1984) 35 – 46. [6] J.H. Dinsmore, F. Solomon, Inhibition of MAP2 expression affects both morphological and cell division phenotypes of neuronal differentiation, Cell 64 (1991) 817 – 826. [7] A. Ferreira, J. Busciglio, C. Landa, A. Caceres, Ganglioside-enhanced neurite outgrowth: evidence for a selective induction of high molecular weight MAP2, J. Neurosci. 10 (1990) 293 – 302. [8] P.W. Frankland, C. O’Brien, M. Ohno, A. Kirkwood, A.J. Silva, Alpha-CaMKII-dependent plasticity in the cortex is required for permanent memory, Nature 411 (2001) 309 – 313. [9] S. Glazewski, C.M. Chen, A. Silva, K. Fox, Requirement for alpha-CaMKII in experience-dependent plasticity of the barrel cortex, Science 272 (1996) 421 – 423. [10] Y. Goshima, S. Ohsako, T. Yamauchi, Overexpression of Ca2+/ calmodulin-dependent protein kinase II in Neuro2a and NG108-15 neuroblastoma cell lines promotes neurite outgrowth and growth cone motility, J. Neurosci. 13 (1993) 559 – 567. [11] H.M. Johnston, B.J. Morris, NMDA and nitric oxide increase microtubule-associated protein 2 gene expression in hippocampal granule cells, J. Neurochem. 63 (1994) 379 – 382. [12] H.M. Johnston, B.J. Morris, N-methyl-D-aspartate and nitric oxide regulate the expression of calcium/calmodulin-dependent kinase II in the hippocampal dentate gyrus, Mol. Brain Res. 31 (1995) 141 – 150. [13] P.M. Lledo, G.O. Hjelmstad, S. Mukherji, T.R. Soderling, R.C. Malenka, R.A. Nicoll, Calcium/calmodulin-dependent kinase II and longterm potentiation enhance synaptic transmission by the same mechanism, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 11175 – 11179. [14] S.E. Mackler, B.P. Brooks, J.H. Eberwine, Stimulus-induced coordinate changes in mRNA abundance in single postsynaptic hippocampal CA1 neurons, Neuron 9 (1992) 539 – 548. [15] B.J. Morris, Neuronal plasticity, in: R.W. Davies, B.J. Morris (Eds.), The Molecular Biology of the Neuron, Oxford University Press, 1996, pp. 323-338. [16] B.J. Morris, Stabilization of dendritic mRNAs by nitric oxide allows localized, activity-dependent enhancement of hippocampal protein synthesis, Eur. J. Neurosci. 9 (1997) 2334 – 2339. [17] D.L. Pettit, S. Perlman, R. Malinow, Potentiated transmission and prevention of further LTP by increased CaMKII activity in postsynaptic hippocampal slice neurons, Science 266 (1994) 1881 – 1885. [18] L.A. Roberts, M.J. Higgins, C.T. O’Shaughnessy, T.W. Stone, B.J. Morris, Changes in hippocampal gene expression associated with the induction of long-term potentiation, Mol. Brain Res. 42 (1996) 123 – 127. [19] L.A. Roberts, C.H. Large, M.J. Higgins, T.W. Stone, C.T. O’Shaughnessy, B.J. Morris, Increased expression of dendritic mRNA following the induction of long-term potentiation, Mol. Brain Res. 56 (1998) 38 – 44. [20] A.J. Silva, C.F. Stevens, S. Tonegawa, Y. Wang, Deficient hippocampal long-term potentiation in a-calcium – calmodulin kinase II mutant mice, Science 257 (1992) 201 – 206. [21] C.S. Simpson, B.J. Morris, Regulation of neuronal cell adhesion molecule expression by NF-kappa B, J. Biol. Chem. 275 (2000) 16879 – 16884. [22] L. Stoppini, A. Buchs, D. Muller, A simple method for organotypic cultures of nervous tissue, J. Neurosci. Methods 37 (1991) 173 – 182. [23] S.J. Tang, G. Reis, H. Kang, A.C. Gingras, N. Sonenberg, E.M. Schuman, A rapamycin-sensitive signaling pathway contributes to
16
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
long-term synaptic plasticity in the hippocampus, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 467 – 472. [24] K.L. Thomas, S. Laroche, M.L. Errington, T.V. Bliss, S.P. Hunt, Spatial and temporal changes in signal transduction pathways during LTP, Neuron 13 (1994) 737 – 745.
[25] L. Wu, D. Wells, J. Tay, D. Mendis, M.A. Abbott, A. Barnitt, E. Quinlan, A. Heynen, J.R. Fallon, J.D. Richter, CPEB-mediated cytoplasmic polyadenylation and the regulation of experience-dependent translation of alpha-CaMKII mRNA at synapses, Neuron 21 (1998) 1129 – 1139.
Research report
Differential regulation of MAP2 and aCamKII expression in hippocampal neurones by forskolin and calcium ionophore treatment K.A. Alier, B.J. Morris * Division of Neuroscience and Biomedical Systems, Institute of Biomedical and Life Sciences, University of Glasgow, West Medical Building, Glasgow G12 8QQ, UK Accepted 19 November 2003
Abstract The genes encoding microtubule-associated protein 2 (MAP2), and the alpha subunit of calcium/calmodulin-dependent protein kinase II (aCaMKII), are members of a small number of genes whose expression is increased in hippocampal neurones during the intermediate phase of long-term potentiation (LTP)—a phase dependent on mRNA translation but not on gene transcription. However, the intracellular signalling pathways which mediate these increases in expression are largely unknown. Organotypic slice cultures of rat hippocampus were exposed to either forskolin (to elevate cAMP levels), A23187 (to increase intracellular Ca2 + levels) or the corresponding vehicle. The levels of immunoreactive (ir-) MAP2 were increased 4 h after forskolin treatment, but were unaffected by A23187 treatment. Conversely, the levels of ir-aCaMKII were increased 4 h after A23187 treatment, but were unaffected by forskolin. The regulation of the expression of these proteins was the same in the CA3 region as in the CA1 and dentate gyrus of the hippocampus. While rapamycin reduced the basal levels of ir-MAP2, it did not affect the ability of either forskolin or A23187 to enhance ir-MAP2 or ir-aCaMKII levels. These results suggest that cAMP and Ca2 + differentially modulate the expression of these two plasticity-related genes, and that translational enhancement via the mammalian target of rapamycin kinase is not involved in these effects. D 2004 Elsevier B.V. All rights reserved. Keywords: Gene expression; Hippocampus; Microtubule-associated protein; Calcium/calmodulin-dependent protein kinase II; cAMP; Rapamycin
1. Introduction Long-term potentiation (LTP) of synaptic transmission is observed after high frequency stimulation of pathways in the hippocampal formation. In the dentate gyrus (DG) and CA1 regions, the induction of LTP is dependent on activation of the NMDA class of glutamate receptor, while in the CA3 region, it is primarily dependent on increased cAMP levels [2]. LTP, in common with most other forms of synaptic plasticity, can be divided into three temporal phases. The earliest phase is sustained by modification of existing proteins, for example, by phosphorylation or subcellular redistribution. The latest phase is sustained by de novo gene transcription and mRNA translation. The intermediate phase, which appears between 1 and 6 h after the initial stimulation, depending on the experimental
* Corresponding author. Tel.: +44-141-330-5361; fax: +44-141-3305659. E-mail address: [email protected] (B.J. Morris). 0169-328X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.molbrainres.2003.11.018
model, is dependent on mRNA translation and de novo protein synthesis, but not on gene transcription [15,16]. There is hence considerable interest in identifying the proteins which are synthesised at this time in response to the stimulation, and the mechanisms which regulate their expression. A handful of mRNAs are present in neuronal dendrites, in the vicinity of the synapses, in addition to the cell soma. Two of these mRNA species, encoding MAP2 and aCaMKII, have been associated with various forms of neuronal plasticity [7– 12,17 –20]. The role of MAP2 in cross-linking microtubules and in facilitating changes in dendritic architecture [3,4,6] makes it an attractive candidate for enabling the synaptic remodelling associated with LTP [5], while CaMKII plays a central role in transducing the effects of NMDA receptor stimulation into lasting intracellular effects [13,17,20]. Increased levels of MAP2 mRNA and aCaMKII mRNA were detected after LTP induction [11,12,14,18,19,24], at least in the dentate gyrus and CA1 regions of the hippocampus. It is not yet known if similar increases are observed
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
in the CA3 region. Current evidence suggests that the increases arise from post-transcriptional regulation of existing mRNA in the dendrites, local to the region of stimulation [16,23]. This local increase in mRNA levels is then translated into elevated levels of the corresponding protein. However, it is unclear what mechanisms couple glutamatergic activation of NMDA receptors, which trigger the induction of LTP, to enhanced MAP2 and aCaMKII mRNA levels. Two major effectors of NMDA receptor activation are Ca2 +, which enters the cell via the activated receptor, and cAMP, which is elevated by the activation of adenyl cyclase. In this paper, we have used hippocampal neurones, maintained in organotypic culture, to study the effect on the levels of MAP2 and aCaMKII of agents which directly increase Ca2 + or cAMP levels. We have tested the hypothesis that Ca2 + influx will be the primary stimulus increasing the expression of both the MAP2 and aCaMKII genes, and that the effects observed in the dentate gyrus and CA1 regions may not be mirrored in the CA3 region.
11
2. Materials and methods 2.1. Cell culture Rat neonates between the ages of 6 and 10 days old were killed using 0.2 ml/animal of an anaesthetic (Euthatal, Rhone Merieux). The hippocampus was removed rapidly and 250 Am hippocampal slices were obtained using a tissue chopper (McIlwain). The slices were then placed in six-well plates (Iwaki) with porous (0.2 Am), transparent, and low protein binding inserts (Falcon) as described [22]. The wells contained 1 ml of Wilde medium each [Basal Medium Eagle (BME) 1 liquid with Earle’s salt without L-glutamine (88.5%), horse serum heat inactivated (17.7%), glucose (22.5 g/l), glutamaxk-1 Supplement (0.46%), penicillin/streptomycin solution (1:106.5 dilutions)]. All the contents of the Wilde medium were purchased from Life Technologies. The six-well plates were then placed in an incubator (95% O2 and 5% CO2) for 4 –5 days.
Fig. 1. The effect of A23187 and forskolin on ir-aCamKII and ir-MAP2 in hippocampal slice cultures. Upper panels: Photomicrographs showing detection of ir-aCamKII after treatment with vehicle (a, b, c) or A23187 (d, e, f), in the dentate gyrus (a, d), CA1 (b, e) and CA3 (c, f) regions. Arrow heads and arrows indicate dendrite and somata, respectively. Note the enhanced staining in all regions after A23187 treatment. Lower panels: Photomicrographs showing detection of ir-MAP2 after treatment with vehicle (a, b, c) or forskolin (d, e, f), in the dentate gyrus (a, d), CA1 (b, e) and CA3 (c, f) regions with the arrow heads and arrows representing the dendrites and somata, respectively. Note the enhanced staining in all regions after forskolin treatment. Scale bar: 20 Am for CA1 and CA3, 15 Am for DG.
12
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
2.2. Hippocampal slice treatments In a first series of experiments, the slices were treated with vehicle [Hanks Balance Salt Solution (HBSS), Life Technologies], forskolin (Sigma, 50 AM) or A23187 (Sigma, 5 AM) in vitro. In a second series of experiments, the slices were treated with vehicle or rapamycin (Calbiochem, 20 nM) for 30 min on the fifth day in vitro before treatment with vehicle, forskolin or A23187. 2.3. Immunocytochemistry Four hours after addition of vehicle, forskolin or A23187, the treated slices were fixed in 3% formaldehyde (Riedel-de Haen) before processing for immunocytochemistry as previously described [16] with monoclonal antibodies specific for aCaMKII (Boehringer, 1:1000) and MAP2 (Sigma, 1:1000) using standard methods. Detection of primary antibody was achieved using biotinylated secondary antibody (anti-mouse) and the ABC (Vector Laboratories) with VIP (Vector Laboratories). The stain-
ing was quantified by using ‘‘Image’’ software (W. Rasband, NIH) as described [21]. The intensity of staining in drug-treated slices was expressed as a percentage of the staining in parallel vehicle-treated slices. The background staining was subtracted to obtain a specific staining using 10 random fields of measurement of different cell bodies/animal. The significance between groups was determined using analysis of variance with post hoc Fisher’s test for multiple pairwise comparison using Minitab statistical program. Significance relative to 100% was assessed by 95% confidence intervals of the mean. 2.4. Western blot Four hours after addition of vehicle, forskolin or A23187, cultured slices were scraped from the inserts and homogenised in RIPA buffer [1 M Tris, pH 8, 1 M NaCl, HP40, 0.5% DOC, 10% SDS, protease inhibitors (Roche Diagnostics)] and centrifuged at 10,000 rpm. After the second centrifugation, the supernatant was denatured and run in NuPAGEk 12% Bis – Tris gels with rainbow
Fig. 2. Semi-quantitative analysis of the effect of A23187 and forskolin on ir-aCamKII in cell bodies (a) or dendrites (b), and ir-MAP2 in cell bodies (c) or dendrites (d) in hippocampal slice cultures. Results are shown for n = 15, *P < 0.05 vs. vehicle treatment ( = 100%) by 95% confidence intervals.
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
13
markers (Amersham Life Science). After blotting into PVDF membrane (Invitrogen), the proteins were visualised with primary antibodies (same as above), HRP secondary antibody (Diagnostics Scotland) and ECL detection kit (Amersham Life Sciences) according to standard procedures. Once the required bands were obtained, the blot was stripped using a stripping buffer [80 mM Tris, pH 6.8, 2% SDS and 68% h-mercaptoethanol (purchased from Sigma)] and reprobed with h-actin (1:5000, Sigma) primary antibody.
3. Results 3.1. The effect of A23187 and forskolin on the expression of ir-aCaMKII and ir-MAP2 Neurones in all hippocampal subregions were stained with the aCaMKII and MAP2 antibodies, with staining prominent in both cell body and dendritic layers. The Ca2 + ionophore, A23187, increased the levels of iraCaMKII in DG, CA3 and CA1 regions of the slice cultures by between 15 –25% in the somata and 9 – 18% in apical dendrites. However, the adenylate cyclase activator, forskolin, did not have any significant effect on iraCaMKII levels in any region (Figs. 1 and 2a,b). Conversely, levels of ir-MAP2 expression were elevated in both cell bodies and dendrites all three regions of the hippocampus in the presence of forskolin while A23187 did not have any significant effect (Figs. 1 and 2c,d). Western blot analysis also indicated an increase in the expression of CaMKII in the presence of A23187 while MAP2 levels were increased in the presence of forskolin (Fig. 3).
Fig. 4. The differential effect of rapamycin on aCaMKII levels in vehicle-, A23187- and forskolin-treated slices: ir-aCaMKII levels were monitored in the cell bodies (a) and dendritic regions (b). Results are expressed as a percentage of the immunocytochemical signal in the corresponding region after vehicle/vehicle treament. ir-aCaMKII levels were not affected by rapamycin pretreatment, but were significantly increased in all three regions of the hippocampus by A23187 (*P < 0.05 versus corresponding region after vehicle treatment in the absence of rapamycin, 95% confidence intervals of the mean). The expression of ir-aCaMKII in DG, CA3 and CA1 was significantly increased in rapamycin/A23187treated slices compared to rapamycin/vehicle ( + P < 0.05, ANOVA with post hoc Fisher test).
3.2. Effect of rapamycin on the expression of ir-aCaMKII and ir-MAP2 in both stimulated and unstimulated hippocampal slices
Fig. 3. The effects of A23187 and forskolin on the hippocampal iraCaMKII and ir-MAP2 were further examined by Western analysis. Bands of predicted sizes were identified (280 and 70 kDa for MAP2, 50 kDa for aCaMKII). The blots were stripped and reprobed with 42 kDa h-actin which was used as an internal standard. The bands for h-actin were uniform, indicating an equal concentration of protein.
The levels of ir-aCaMKII in cell bodies and dendrites were not significantly affected by pretreatment of slices with rapamycin in any of the hippocampal regions (Fig. 4a,b). For A23187-treated hippocampal slices, there was a significant increase in the levels of ir-aCaMKII in both vehicletreated and rapamycin-pretreated slices. This increase occurred in DG, CA1 and CA3 regions (Fig. 4a,b). This indicated that rapamycin had no effect on the elevation of ir-aCaMKII induced by A23187. There was a significant decrease in the basal levels irMAP2 after pretreatment of slices with rapamycin in the
14
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
Fig. 5. The differential effect of rapamycin on MAP2 levels in vehicle-, A23187- and forskolin-treated slices: ir-MAP2 levels were monitored in the cell bodies (a) and dendritic regions (b). Results are expressed as a percentage of the immunocytochemical signal in the corresponding region after vehicle/vehicle treament. The basal expression of ir-MAP2 was significantly reduced by rapamycin in all three regions of the hippocampus (*P < 0.05 versus vehicle treatment without rapamycin, 95% confidence intervals of the mean), ir-MAP2 levels were significantly increased in all three regions of the hippocampus by forskolin (*P < 0.05 versus corresponding region after vehicle treatment in the absence of rapamycin, 95% confidence intervals of the mean). The expression of ir-MAP2 in DG, CA3 and CA1 was significantly increased in rapamycin/forskolin-treated slices compared to rapamycin/vehicle ( f P < 0.05, ANOVA with post hoc Fisher test).
somata (Fig. 5a) compared to slices receiving vehicle pretreatment. Forskolin treatment increased ir-MAP2 levels (in somata and dendrites) relative to vehicle treatment, even after pretreatment with rapamycin, indicating that rapamycin had no effect on the elevation of ir-MAP2 in all three regions of the hippocampus during stimulation (Fig. 5).
4. Discussion The results indicate that treatment with A23187 increases hippocampal ir-aCaMKII while forskolin has
no effect. This suggests the importance of Ca2 + influx in regulating the expression of aCaMKII. One possible mechanism by which Ca2 + influx stimulates aCaMKII synthesis is via phosphorylation of cytoplasmic polyadenylation element binding protein (CPEB), which is present in the hippocampal dendrites and somata [25]. Wu et al. [25] provided evidence that binding of CPEB to CPE sites located in the 3V-end of the RNA message for aCaMKII can stimulate its translation rate. However, as yet there is no evidence that CPEB is phosphorylated in response to Ca2 + influx. The Ca2 + ionophore (A23187) does not produce any significant effect on ir-MAP2 levels. However, elevation of ir-MAP2 levels is observed after forskolin treatment, suggesting the involvement of cAMP in the regulation of MAP2 expression. Forskolin activates adenylate cyclase (AC), which increases the levels of cAMP. AC2 and AC4 have been shown to be expressed in the hippocampal formation and are co-localised with MAP2 [1]. AC2 and AC4 have been labelled in the dendrites and cell bodies of DG, CA3 and CA1 [1]. Hence these two isoforms are likely to be involved in the expression of ir MAP2. Therefore cAMP and PKA might act on the MAP2 mRNA via an unknown mechanism resulting in the elevation of MAP2 in the three regions of the hippocampus. These results provide strong evidence that the elevation of ir aCaMKII and MAP2 associated with synaptic plasticity may be achieved via separate pathways. This in turn would suggest that the two genes are not necessarily regulated in parallel, but could show a differential relative change in expression according to the relative change in Ca2 + and cAMP levels. The data also suggests that the elevation of mRNA in the dendrites plays a role in the expression of these proteins in the dendritic region of the hippocampus. The levels of the two proteins were elevated in all the three regions of the hippocampus i.e. DG, CA3 and CA1, suggesting that the intracellular pathways regulating MAP2 and aCaMKII expression in CA3 are similar to those in CA1 and DG. This is the first evidence that CA3 neurones can show enhanced MAP2 and CaMKII expression in the same way as DG and CA1 neurones. Rapamycin produces its effect by binding to and inhibiting a kinase, the mammalian target of rapamycin (mTOR). It is well known that mTOR can activate mRNA translation via phosphorylation of ribosomal accessory proteins. Rapamycin hence acts to switch off translation resulting in protein synthesis inhibition. Rapamycin reportedly inhibits the intermediate phase of LTP [23], and so mTOR could be involved in stimulating translation of proteins such as MAP2 and aCaMKII. Rapamycin did not have any significant effect on either the basal expression of aCaMKII (Fig. 4) or the increased expression after A23187 treatment. Hence the regulation of aCaMKII expression appears to be achieved via a rapamycin-insensitive pathway. However, rapamycin decreased the basal expression of ir-MAP2 significantly, but did not have any effect on ir-MAP2 stimulation by forskolin (Fig. 5),
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
suggesting that at basal levels inhibition of MAP2 expression is achieved via a rapamycin-sensitive pathway while after stimulation the change in expression occurs through a rapamycin-insensitive pathway. This suggests that aCaMKII CaMKII and MAP2 are not among any rapamycin-sensitive genes involved in the intermediate phase of LTP, and is consistent with the concept that post-translational activation of these genes occurs through other pathways, such as mRNA stabilisation [16]. Our initial hypothesis—that Ca2 + influx would be the primary stimulus increasing the expression of both the MAP2 and aCaMKII genes, and that the effects observed in the dentate gyrus and CA1 regions might not be mirrored in the CA3 region—therefore proved over-simplistic. Clearly Ca2 + influx is not the primary stimulus regulating all plasticity-related genes, and cAMP can play a major role distinct from Ca2 + influx. Furthermore, we found the regulation of these genes to be identical in the CA3 region as compared to the dentate gyrus and CA1 region. Hence while the mechanisms for triggering the plasticity response are distinct in the CA3 region, the mechanisms regulating the expression of late-response genes appear to be the same. This implies that there may be many commonalities in the genomic changes in different models of late-phase synaptic plasticity.
5. Conclusion Ca2 + plays an important role in the expression of iraCaMKII while an increase in the levels of cAMP affects the levels of expression of ir-MAP2. Intracellular pathways for CA3 are similar to DG and CA1 since the levels of both proteins were elevated in all the three regions of the hippocampus. Basal levels of ir-MAP2 follow the rapamycin-sensitive pathway while the stimulated levels of ir-MAP2 and both basal and stimulated levels ir-aCaMKII follow the rapamycin-insensitive pathway, suggesting different post-translational activation pathways.
Acknowledgements This work would not have been possible without the support of the Africa Educational Trust (AET)—we are very grateful for their support.
References [1] L.P. Baker, M.D. Nielsen, S. Impey, B.M. Hacker, S.W. Poser, M.Y. Chan, D.R. Storm, Regulation and immunohistochemical localization of betagamma-stimulated adenylyl cyclases in mouse hippocampus, J. Neurosci. 19 (1999) 180 – 192. [2] T.V.P. Bliss, G.L. Collingridge, A synaptic model of memory: longterm potentiation in the hippocampus, Nature 361 (1993) 31 – 38.
15
[3] A. Caceres, J. Busciglio, A. Ferreira, O. Steward, An immunocytochemical and biochemical study of the microtubule-associated protein MAP-2 during post-lesion dendritic remodeling in the central nervous system of adult rats, Brain Res. 427 (1988) 233 – 246. [4] A. Caceres, J. Mautino, K.S. Kosik, Suppression of MAP2 in cultured cerebellar macroneurons inhibits minor neurite formation, Neuron 9 (1992) 607 – 618. [5] F.L. Chang, W.T. Greenough, Transient and enduring morphological correlates of synaptic activity and efficacy change in the rat hippocampal slice, Brain Res. 309 (1984) 35 – 46. [6] J.H. Dinsmore, F. Solomon, Inhibition of MAP2 expression affects both morphological and cell division phenotypes of neuronal differentiation, Cell 64 (1991) 817 – 826. [7] A. Ferreira, J. Busciglio, C. Landa, A. Caceres, Ganglioside-enhanced neurite outgrowth: evidence for a selective induction of high molecular weight MAP2, J. Neurosci. 10 (1990) 293 – 302. [8] P.W. Frankland, C. O’Brien, M. Ohno, A. Kirkwood, A.J. Silva, Alpha-CaMKII-dependent plasticity in the cortex is required for permanent memory, Nature 411 (2001) 309 – 313. [9] S. Glazewski, C.M. Chen, A. Silva, K. Fox, Requirement for alpha-CaMKII in experience-dependent plasticity of the barrel cortex, Science 272 (1996) 421 – 423. [10] Y. Goshima, S. Ohsako, T. Yamauchi, Overexpression of Ca2+/ calmodulin-dependent protein kinase II in Neuro2a and NG108-15 neuroblastoma cell lines promotes neurite outgrowth and growth cone motility, J. Neurosci. 13 (1993) 559 – 567. [11] H.M. Johnston, B.J. Morris, NMDA and nitric oxide increase microtubule-associated protein 2 gene expression in hippocampal granule cells, J. Neurochem. 63 (1994) 379 – 382. [12] H.M. Johnston, B.J. Morris, N-methyl-D-aspartate and nitric oxide regulate the expression of calcium/calmodulin-dependent kinase II in the hippocampal dentate gyrus, Mol. Brain Res. 31 (1995) 141 – 150. [13] P.M. Lledo, G.O. Hjelmstad, S. Mukherji, T.R. Soderling, R.C. Malenka, R.A. Nicoll, Calcium/calmodulin-dependent kinase II and longterm potentiation enhance synaptic transmission by the same mechanism, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 11175 – 11179. [14] S.E. Mackler, B.P. Brooks, J.H. Eberwine, Stimulus-induced coordinate changes in mRNA abundance in single postsynaptic hippocampal CA1 neurons, Neuron 9 (1992) 539 – 548. [15] B.J. Morris, Neuronal plasticity, in: R.W. Davies, B.J. Morris (Eds.), The Molecular Biology of the Neuron, Oxford University Press, 1996, pp. 323-338. [16] B.J. Morris, Stabilization of dendritic mRNAs by nitric oxide allows localized, activity-dependent enhancement of hippocampal protein synthesis, Eur. J. Neurosci. 9 (1997) 2334 – 2339. [17] D.L. Pettit, S. Perlman, R. Malinow, Potentiated transmission and prevention of further LTP by increased CaMKII activity in postsynaptic hippocampal slice neurons, Science 266 (1994) 1881 – 1885. [18] L.A. Roberts, M.J. Higgins, C.T. O’Shaughnessy, T.W. Stone, B.J. Morris, Changes in hippocampal gene expression associated with the induction of long-term potentiation, Mol. Brain Res. 42 (1996) 123 – 127. [19] L.A. Roberts, C.H. Large, M.J. Higgins, T.W. Stone, C.T. O’Shaughnessy, B.J. Morris, Increased expression of dendritic mRNA following the induction of long-term potentiation, Mol. Brain Res. 56 (1998) 38 – 44. [20] A.J. Silva, C.F. Stevens, S. Tonegawa, Y. Wang, Deficient hippocampal long-term potentiation in a-calcium – calmodulin kinase II mutant mice, Science 257 (1992) 201 – 206. [21] C.S. Simpson, B.J. Morris, Regulation of neuronal cell adhesion molecule expression by NF-kappa B, J. Biol. Chem. 275 (2000) 16879 – 16884. [22] L. Stoppini, A. Buchs, D. Muller, A simple method for organotypic cultures of nervous tissue, J. Neurosci. Methods 37 (1991) 173 – 182. [23] S.J. Tang, G. Reis, H. Kang, A.C. Gingras, N. Sonenberg, E.M. Schuman, A rapamycin-sensitive signaling pathway contributes to
16
K.A. Alier, B.J. Morris / Molecular Brain Research 122 (2004) 10–16
long-term synaptic plasticity in the hippocampus, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 467 – 472. [24] K.L. Thomas, S. Laroche, M.L. Errington, T.V. Bliss, S.P. Hunt, Spatial and temporal changes in signal transduction pathways during LTP, Neuron 13 (1994) 737 – 745.
[25] L. Wu, D. Wells, J. Tay, D. Mendis, M.A. Abbott, A. Barnitt, E. Quinlan, A. Heynen, J.R. Fallon, J.D. Richter, CPEB-mediated cytoplasmic polyadenylation and the regulation of experience-dependent translation of alpha-CaMKII mRNA at synapses, Neuron 21 (1998) 1129 – 1139.