Quantitative Proteomics and Protein Network Analysis of Hippocampal

Quantitative Proteomics and Protein Network Analysis of Hippocampal Synapses of CaMKIIr Mutant Mice Ka Wan Li,*,† Stephan Miller,# Oleg Klychnikov,† Maarten Loos,† Jianru Stahl-Zeng,§ Sabine Spijker,† Mark Mayford,# and August B. Smit† Department of Molecular and Cellular Neurobiology, Center for Neurogenomics and Cognitive Research, Faculty of Earth and Life Sciences, Vrije Universiteit, Amsterdam, The Netherlands, Department of Cell Biology and Institute for Childhood and Neglected Diseases, The Scripps Research Institute, La Jolla, California, and Applied Biosystems, Darmstadt, Germany Received February 16, 2007

Quantitative analysis of synaptic proteomes from specific brain regions is important for our understanding of the molecular basis of neuroplasticity and brain disorders. In the present study we have optimized comparative synaptic proteome analysis to quantitate proteins of the synaptic membrane fraction isolated from the hippocampus of wild type mice and 3′UTR-calcium/calmodulin-dependent kinase II R mutant mice. Synaptic proteins were solubilized in 0.85% RapiGest and digested with trypsin without prior dilution of the detergent, and the peptides from two groups of wild type mice and two groups of CaMKIIR 3′UTR mutants were tagged with iTRAQ reagents 114, 115, 116, and 117, respectively. The experiment was repeated once with independent biological replicates. Peptides were fractionated with tandem liquid chromatography and collected off-line onto MALDI metal plates. The first iTRAQ experiment was analyzed on an ABI 4700 proteomics analyzer, and the second experiment was analyzed on an ABI 4800 proteomics analyzer. Using the criteria that the proteins should be matched with at least three peptides with the highest CI% of a peptide at least 95%, 623 and 259 proteins were quantified by a 4800 proteomics analyzer and a 4700 proteomics analyzer, respectively, from which 249 proteins overlapped in the two experiments. There was a 3 fold decrease of calcium/calmodulin-dependent kinase II R in the synaptic membrane fraction of the 3′UTR mutant mice. No other major changes were observed, suggesting that the synapse protein constituents of the mutant mice were not substantially altered. A first draft of a synaptic protein interaction network has been constructed using commercial available software, and the synaptic proteins were organized into 10 (interconnecting) functional groups belonging to the pre- and postsynaptic compartments, e.g., receptors and ion channels, scaffolding proteins, cytoskeletal proteins, signaling proteins, adhesion molecules, and proteins of synaptic vesicles and those involved in membrane recycling. Keywords: ITRAQ • mutant mice • synaptic proteome • hippocampus • protein network

Introduction The brain is the most complex and dynamically organized organ with a high degree of computation capability that enables acquisition and integration of information in order to generate appropriate responses to environmental and physiological inputs. Pivotal to its function is the extensive neuronal connectivity, in particular the 1015 synaptic connections of the 100 billion central neurons of the human brain, via which neurotransmission occurs. Synaptic transmission involves the presynaptic release of transmitters and the activation of transmitter receptors and signal transduction cascades in the postsynaptic compartment. Synaptic transmission can be * Author to whom correspondence should be addressed. Telephone: 3120-5987107. Fax: 31-20-5989281. E-mail: [email protected]. † Vrije Universiteit. # The Scripps Research Institute. § Applied Biosystems. 10.1021/pr070086w CCC: $37.00

 2007 American Chemical Society

strengthened or weakened.1 These changes in synaptic strength are at the basis of information storage through which higher order brain functioning are thought to be established. The underlying molecular mechanisms of synaptic plasticity underpin the functional characteristics of the brain regions involved, e.g., molecular alterations in synaptic proteins have been shown to underlie hippocampal long-term potentiation/ long-term depression and are involved in learning and memory formation. Also, abnormal alterations of synaptic transmission are thought to be associated with a number of brain disorders, such as drug addiction2 and schizophrenia.3 A primary event that underlies plasticity of the synapse is the calcium influx via the activity-dependent NMDA receptor. Ca-influx results in the activation of downstream signaling cascades, notably the calcium/calmodulin-dependent kinase II R (CaMKIIR) dependent transduction pathway.4 Sustained activation of the synapse causes trafficking of CaMKIIR to the Journal of Proteome Research 2007, 6, 3127-3133

3127

Published on Web 07/11/2007

research articles postsynaptic density, the subsynaptic compartment where transduction of neurotransmitter receptor-mediated signaling takes place. The activation and translocation of CaMKIIR to the postsynaptic density is thought to be the primary event in the initiation of synaptic plasticity,5 leading to long-term adjustments in efficacy of synaptic transmission important for learning and memory formation. Recently, others and we have used various proteomics approaches to examine protein constituents of synaptic structures including presynaptic zone, postsynaptic density, and synaptic membrane.6-20 To facilitate protein characterization in the synapse large amounts of starting material from multiple brain regions were used. Notwithstanding the usefulness of these studies the results are mostly descriptive in nature. To get a better insight into synapse physiology it will be important to understand the organization of proteins into functional groups within a particular brain region(s), in which they execute distinct actions, and also how they are regulated under conditions of different neuronal activity. Considering the vast number of clinical and fundamental studies on synaptic functioning it is highly desirable to develop a simple quantitative proteomics analysis in order to examine and compare larger numbers of samples of selected brain regions. This is technically challenging because quantification needs to be carried out from a low amount of starting material that is rich in membrane proteins.21 Furthermore, modeling of a synaptic protein interaction network is in its infancy. In this study we have optimized an iTRAQ-based labeling technique22-24 in conjunction with tandem liquid chromatography-MS/MS for protein characterization and quantification from a synaptic membrane preparation. We have compared the synaptic proteomes isolated from hippocampus of the wild type mice (WT) and of CaMKIIR mutant mice (3′UTR mutant). In this mutant mouse, the dendritic localization signal in the mRNA of CaMKIIR in the 3′ untranslated region is disrupted so that mRNA is restricted to the soma.25 These mice exhibit impairment of events that require plastic changes of the synapse in hippocampus, such as late-phase long-term potentiation, spatial memory, associative fear conditioning, and object recognition memory. We detected 3 fold decrease of CaMKIIR in the synaptic membrane of 3′UTR mutant. Interestingly, the overall synaptic membrane proteome was not altered. Finally, we have constructed a draft of a synaptic protein interaction network using existing pathways analysis software.26

Experimental Methods Synaptic Membrane Preparation. Synaptic membranes were isolated as described previously.10,14 In brief, hippocampi from five mice were homogenized in 15 mL of 320 mM sucrose and then centrifuged at 1000g for 10 min. Four samples were processed simultaneously, i.e., two groups of WT and two groups of 3′UTR mutant. Supernatant was loaded on top of a sucrose step gradient consisting of 0.85 and 1.2 M sucrose. After centrifugation at 100000g for 2 h the synaptosome fraction at the interface of 0.85/1.2 M sucrose was collected and then lysed in hypotonic solution. The resulting synaptic membrane fraction was recovered by centrifugation using the sucrose step gradient as stated above. The synaptic membrane fraction was collected from the 0.85/1.2 M interface and dried on a SpeedVac. The experiment was repeated once with another set of two groups of WT and two groups of 3′UTR mutant. Protein Digestion and iTRAQ Labeling. Synaptic membrane fraction proteins were resuspended in 28 µL of dissolution 3128

Journal of Proteome Research • Vol. 6, No. 8, 2007

Li et al.

buffer and 2 µL of cleavage reagent (iTRAQ reagent kit, Applied Biosystems) with 0.85% RapiGest (Waters associates) to solubilize membrane proteins. After incubation for 1 h, 1 µL of cys blocking reagent was added and vortexed for 20 min. Subsequently, trypsin dissolved in 10 µL of water was added and incubated overnight at 37 °C. As trypsin remains active at even high concentration of RapiGest, no prior dilution of the sample to reduce detergent concentration was necessary, and the final sample volume was maintained at around 40 µL. The trypsinized peptides were then tagged with iTRAQ reagents dissolved in 85 µL of ethanol. The two WT groups were labeled respectively with iTRAQ reagents 114 and 115; the two 3′UTR mutant groups with iTRAQ reagents 116 and 117. After incubation for 3 h the samples were pooled together and acidified with 10% trifluoroacetic acid to pH 2.5-3. The low pH cleaves RapiGest which can then be removed by centrifugation. After 1 h, the sample was centrifuged and the supernatant dried on a SpeedVac. Two-Dimensional Liquid Chromatography. The dried iTRAQlabeled sample was dissolved in 300 µL of loading buffer (20% acetonitrile, 10 mM KH2PO4, pH 2.9) and loaded into a polysulfoethyl A column (PolyLC). Peptides were eluted with a linear gradient of 0-500 mM KCl in 20% acetonitrile, 10 mM KH2PO4, pH 2.9, over 25 min at a flow rate of 50 µL/min. Fractions were collected at 1-min intervals. In the second dimensional liquid chromatography separation, peptides were delivered with a FAMOS autosampler at 30 µL/min to a C18 trap column (1 mm × 300 µm i.d. column) and separated on an analytical capillary C18 column (150 mm × 100 µm i.d. column) at 500 nL/min using the LC-Packing Ultimate system. The peptides were separated using a linearly increasing concentration of acetonitrile from 5 to 50% in 30 min, and to 100% in 5 min. The eluent was mixed with matrix (7 mg of R-cyanohydroxycinnaminic acid in 1 mL of 50% acetronitrile, 0.1% trifluoroacetic acid, 10 mM ammonium dicitrate) delivered at a flow rate of 1.5 µL/min, and deposited off-line to the Applied Biosystems metal target every 15 s for a total of 192 spots, using an automatic robot, the Probot from Dionex. Mass Spectrometry and Data Analysis. There were two sample sets, each set containing the iTRAQ-labeled synaptic membrane proteins from two groups of WT and two groups of 3′UTR mutant. The first set of sample was analyzed on an ABI 4700 proteomics analyzer,10,14 whereas the second set of sample was analyzed on an ABI 4800 proteomics analyzer. Peptide CID was performed at 1 kV, and the collision gas was either nitrogen or air. MS/MS spectra were collected from 5000 laser shots in 4700 proteomics analyzer and 2500 laser shots in 4800 proteomics analyzer. The peptides with signal-to-noise ratio above 50 at the MS mode were selected for MS/ MS experiment; a maximum of 30 MS/MS was allowed per spot. The precursor mass window was 180 relative resolution (fwhm). MS/MS spectra were searched against mice database using GPS Explorer (ABI) and Mascot (MatrixScience) with trypsin specificity and fixed iTRAQ modifications at lysine residue and the N-termini of the peptides. Mass tolerance was 100 ppm for precursor ions and 0.5 Da for fragment ions; one missed cleavage was allowed. For each MS/MS spectrum, a single peptide hit with the highest Mascot score in the Swissprot database was considered for further analysis. If a spectrum could not be annotated using Swissprot database, a second Mascot search was performed in the larger but more redundant NCBI database. Next, the precursor protein sequences of all peptides were retrieved from the respective databases, and NCBI sequence that shared more

Quantitative Synaptic Proteomics

than 90% similarity over 85% of the sequence length with a Swissprot sequence were clustered together as a single protein cluster. Finally, all peptides were matched against the protein clusters; those that were matched to more than one protein represent common protein motifs and were not considered for protein identification and quantification. Peak areas for each iTRAQ signature peak (m/z 114.1, 115.1, 116.1, and 117.1) were obtained and corrected according to the manufacturers’ instruction to account for the differences in isotopic overlap. To compensate for the possible variations in the starting amounts of the samples, the individual peak areas of each iTRAQ signature peak were log2 transformed to obtain a normal distribution (data not shown) and then normalized to the total peak area of this signature peak. Low signature peaks generally have larger variation which may compromise the quantitative analysis of the proteins. Therefore, iTRAQ signature peaks less than 2000 were removed from quantitation. For each peptide, the mean peak area of the WT and 3′UTR mutant samples were used to calculate the regulation (KO-WT; log2-scale). Within an experiment average regulation, standard deviation and p-value (two-tailed Student’s t-test) were calculated over four mutant samples versus four WT samples in the two experiments. Protein Interaction Network Modeling. The files were uploaded to the server of “Ingenuity Pathways Analysis” (IPA) for the construction of a protein interaction network. Two separate protein lists were uploaded. First, to construct the synapse protein interaction network we used the combined proteins from the two experiments that were identified by at least three peptides, and at least one of them with CI% >95. IPA identifies genes associated with a list of relevant functions and diseases, and the list is ranked according to the significance of the biological functions. The associated genes to the major biological functions recognized by the IPA from our dataset were selected and added to a pathway. As only 275 proteins can be included in a single protein network, we focused on synapse specific proteins and deleted proteins involved in energy production and carbohydrate metabolism from the list. We used direct interactions, and all the types of relationship. The proteins were then clustered manually according to their cellular localization or gene ontology to visualize the spatial and functional interactions of the subnetworks. Second, for comparative analysis we considered only the overlapping proteins from the two experiments that were identified by at least three peptides in both experiments. The degree of significance of protein expression difference between WT and 3′UTR mutant overlays the synapse protein interaction network. Western Blotting. Three independent biological replicates of synaptosomes were prepared from hippocampi of WT and 3′UTR mutant. Synaptosomes (10 µg) were mixed with 2× SDS sample buffer and boiled for 5 min. Proteins were separated by SDS-PAGE and transferred onto nitrocellulose membrane. Membranes were incubated with antibodies against CaMKIIR (Zymed), CaMKIIβ (Zymed), and NMDA receptor (BD pharrmingen), followed by an alkaline phosphatase-conjugated antibody (Dako) and an enhanced chemifluoresence substrate. The image was scanned with a FLA 5000 (Fujifilm) and analyzed with Quantity One (BioRad).

Results In order to specifically examine synaptic proteins, a synaptic membrane fraction was isolated from the hippocampus of WT and the 3′ UTR mutant. A characteristic of the synaptic

research articles membrane is its high lipid content and high density of membrane proteins. Because iTRAQ labeling should be performed on a preparation solubilized in preferably less than 40 µL of aqueous buffer, from several detergents tested (data not shown) the acid labile ionic detergent, RapiGest, turns out to be most useful (solubilization in 28 µL of 0.85% RapiGest). Trypsination and iTRAQ labeling were then performed. Peptides were separated by ion exchange and nanoC18 column chromatography and used for MS/MS analysis. Four biological replicates of WT and 3′UTR mutant were used in two iTRAQ experiments. For each experiment around 20 000 MS/MS were performed on the pooled iTRAQ-labeled synaptic membrane preparation of WT and 3′UTR mutant, from which 1122 and 495 distinct proteins were identified using 4800 and 4700 proteomics analyzer, respectively. The higher number of proteins identified by 4800 proteomics analyzer is consistent with the expected increase in sensitivity of this new generation of MALDI TOF/TOF MS. To obtain a more reliable quantitative analysis we adopted the stringent selection criterion in which at least three peptides were matched per protein. In all cases the CI% for at least one of the peptides was higher than 95. The iTRAQ peak area must be at least 2000. Proteins expressed at low level most likely did not pass this selection filter and were excluded from the analysis. In total, 623 and 259 proteins were quantified by 4800 proteomics analyzer and 4700 proteomics analyzer, respectively, from which 249 proteins overlapped in the two experiments (Supporting Information; Table 1). To get a better insight into the structural and functional organization of these proteins in the synapse, we used IPA to construct a synaptic protein interaction network. Proteins identified with at least three peptides from at least one experiment were imported into IPA. IPA analysis classifies the biological system as highly related to cell-to-cell signaling and interaction, nervous system development and function, cellular function and maintenance, molecular transport, etc. As a maximum of 275 proteins are allowed to construct a protein interaction map in IPA, we further focused on specific synaptic proteins, and deleted proteins that are involved in carbohydrate metabolism and energy production. Figure 1 shows the interaction of glutamate receptors (A) that are connected to the scaffolding proteins (B) and eventually to the cytoskeletal proteins (C). The signaling proteins, especially kinases and phosphatases, are clustered in D. A large number of proteins are involved in actin dynamics (E). The postsynaptic compartment is anchored within the synapse via actions of adhesion molecules (H). G proteins and G-protein coupled receptors commonly occur in the synaptic membrane (I). In the presynaptic compartment there are many synaptic vesicle proteins and proteins involved in vesicle release (F), and the molecular machinery involved in endocytotic event (G) that is needed to retrieve the synaptic vesicle membrane from the active zone. Furthermore, there are a number of small G proteins, Rabs, that are implicated in vesicle trafficking. Several ion channels including potassium, sodium, and calcium channels and GABAA receptors are grouped together in J. Members of solute carrier family including glutamate and GABA transporters, as well as members of ATPase for transport of ions, are grouped loosely together at the upper right-hand corner of the model. The degree of significance in protein expression between WT and 3′UTR mutant are represented by the intensity of yellow color. Proteins that were identified by single experiment are not considered for quantification. As expected, the expression Journal of Proteome Research • Vol. 6, No. 8, 2007 3129

research articles

Li et al.

Figure 1. Hippocampal synaptic proteins network. Proteins identified with three peptides were imported into IPA, which classifies the proteins into several functional groups. Specific synaptic proteins were selected for network modeling, while proteins that are involved in carbohydrate metabolism and energy production were excluded from the analysis. The proteins are clustered manually according to their cellular localization and/or gene ontology into 10 groups; (A) glutamate receptors complex, (B) scaffolding proteins, (C) cytoskeletal proteins, (D) signaling proteins, especially kinases and phosphatases, (E) proteins that are involved in actin dynamics, (F) synaptic vesicle proteins and proteins involved in vesicle release, (G) molecular machinery involved in endocytotic event, (H) adhesion molecules, (I) G proteins and G-protein coupled receptors, and (J) several ion channels including potassium, sodium and calcium channels and GABAA receptors. The degree of significance in protein expression difference between WT and 3′UTR mutant are represented by the intensity of yellow color.

difference of CaMKIIR is highly significant, and with a fold change of about 3 (Supporting Information; Table 1). Other proteins with good significance are less clear-cut because the fold changes were either small (voltage-dependent anionselective channel protein 3, protein kinase C gamma, and guanine nucleotide-binding protein G; 0.05).

Quantitative Synaptic Proteomics

research articles

Figure 2. iTRAQ ratios of individual identified peptides from representative proteins. X-axis, amino acid sequences of each identified peptides matched to the corresponding proteins. Y-axis, deviations of iTRAQ ratio from average, log2-scale. In order to plot peptides in the same graph, peak areas were rescaled by subtracting the mean of the peak areas of the four samples. Red represents WT, and blue the 3′UTR mutant mice.

groups, within the synapse has been constructed and visualized by a commercially available and easy to use pathways analysis software, which aids the interpretation of the proteomics data.26,27

Figure 3. Western blotting analyses confirm the change of CaMKIIR expression in 3′UTR mutant. Three independent biological replicates of synaptosomes from WT and 3′UTR mutant were used. The CaMKIIR level of 3′UTR mutant (Mu) is about 20% of the WT (Wt) mice. The other two proteins, CaMKIIβ and NMDA receptor, do not show difference between 3′UTR mutant and WT.

Discussion We have developed a straightforward approach for quantitative analysis of synaptic proteomes from mouse brain. This approach will also aid the identification of drug and therapeutic targets of brain diseases. Specifically, we examined protein expression changes in the synapse of hippocampus isolated from the CaMKIIR 3′UTR mutant. This 3′UTR mutant is known to have reduced levels of CaMKIIR due to lack of dendritic synthesis of the protein. Our data are consistent with earlier findings obtained by western blotting that these mice showed a dramatic reduction in CaMKIIR levels in postsynaptic density.25 The protein interaction sets, and thus the functional

In this study we used five hippocampi per sample, but good results were also obtained using two hippocampi per sample (data not shown). The first challenge of the study was to solublize the synaptic proteome. A large number of synaptic proteins, especially those localized in the postsynaptic density, are insoluble in most solvents including non-ionic detergents, such as Triton X-100.10,14 SDS is generally used in this respect, but it is incompatible with trypsin digestion unless the SDS concentration is diluted to less than 0.1%. Many-fold dilution of the solubilization solvent is not favorable for iTRAQ labeling because the reagent will be autohydrolyzed in aqueous environment. In the newly available detergent, RapiGest, 30 µL of 0.85% RapiGest was sufficient to completely solubilize the synaptic membranes isolated from two to five hippocampi. Significantly, trypsin activity is not impaired at this detergent concentration. During the data analysis it became clear that there is a considerable redundancy in NCBI accession numbers. This is a drawback for appropriate protein identification and potentially interferes with the quantification of proteins. To circumvent these problems we have first searched the nonredundant Swissprot database. If a spectrum could not be annotated using Swissprot database, a second Mascot search was performed using the larger but more redundant NCBI database. Generally Journal of Proteome Research • Vol. 6, No. 8, 2007 3131

research articles this second round of database search retrieves another 10% of proteins. Next, the precursor protein sequences of all peptides were retrieved from the respective databases, and NCBI sequences that shared more than 90% similarity over 85% of the sequence length with a Swissprot sequence were grouped together as a single protein cluster. For proteins belonging to the same family they may contain conserved domains and therefore may share peptide identity. Because it is impossible to assign these peptides to single proteins, we excluded them from protein quantification. As a result, a nonredundant synaptic proteome was constructed. Undoubtedly, some of these proteins may be false positives. To remove false positives and to improve quantitation we only considered proteins that were matched with three or more peptides and with CI% above 95% for the highest scored peptide. Consequently, the low expressed proteins in the synapse will be excluded. Nevertheless, about 300-700 proteins were retrieved and were used for the construction of synaptic protein interaction network. We have tested several pathways analysis software packages, and have chosen IPA for further data analysis, mainly for two reasons. First, all protein interactions in the IPA database are manually curated. Second, the networks generated by IPA are displayed orderly to functional groups and can be easily interrogated. The Ingenuity Pathways Knowledge Base can establish interaction between Focus Genes and all other genes stored in the knowledge base, which generates a set of small networks containing a maximum of 35 protein nodes each. To construct edges among the Focus Genes, extra proteins may be added from Ingenuity Pathways Knowledge Base automatically. Alternatively, as IPA identifies genes that are associated with a list of relevant functions and diseases, which are in turn ranked according to the significance of that biological function, it is possible to select all the associated genes and construct a protein interaction network. We have chosen the latter approach because it gives us a global view of the network for a set of not manipulated data. We have enabled direct interactions, and included all the relationship types. The proteins are then clustered according to their cellular localization or gene ontology to visualize the spatial and functional interactions in the subgraphs. This generates a draft of a hippocampus synaptic protein interaction map. Some proteins within the subgraphs exist as single nodes with no edge. It is likely that the interaction patterns of these proteins are not yet present in the IPA database. Alternatively, the interacting partners may not be included in our analysis set because they may have been filtered off from our stringent criterion or have been missed from our MS/MS analysis. To define individual protein interactions within the network of the synapse,19 large-scale proteomics analysis of protein complexes will be needed. There is a general consent that quantitative synaptic proteomics is important for studies of the healthy and diseased brain and for studies on different animal models. Many descriptive synaptic proteomics have been reported.6-20 Recently, iTRAQ quantitation has been applied to examine the cerebellar protein changes in mice that are deficient in plasma membrane calcium ATPase 228 and the protein abundance levels in brain tissue afflicted with AD relative to normal brain tissue.22 Here we demonstrate that iTRAQ labeling in conjunction with LC-LC MS/MS analysis is a useful approach for quantitative analysis of synaptic proteomes. We assessed the hippocampal synaptic proteomes of CaMKIIR 3′UTR mutant mice, that, because of disruption of dendritic localization of 3132

Journal of Proteome Research • Vol. 6, No. 8, 2007

Li et al.

the CaMKIIR mRNA, cannot de novo synthesize CaMKIIR locally in the synapse25 and compared these with WT mice. CaMKIIR is a key regulatory kinase that modulates synaptic strength29 by phosphorylation of various substrate proteins causing changes of protein function and protein-protein interactions, e.g., causing the recruitment of receptor proteins to the postsynaptic density.30 Previously, it was demonstrated that the CaMKIIR 3′UTR mutant mice display a dramatic reduction of CaMKIIR levels in postsynaptic densities. Interestingly, the synaptic efficacy, short-term plasticity and early forms of long-term potentiation are not affected in the 3′UTR mutant suggesting that at basal levels of activity these synapses function normally.25 In the present study, we showed that CaMKIIR is 3 fold depleted in the 3′UTR mutant. In agreement with the synapse physiology, the synaptic proteomes of WT and 3′UTR mutant are quite similar. The major differences between 3′UTR mutant and WT mice are the events that require plastic changes of the synapse, including reduction in late-phase long-term potentiation, impairments in spatial memory, associative fear conditioning, and object recognition memory.25 Future studies will be focused on the synaptic proteomes of 3′UTR mutant and WT mice during the expression and/or maintenance phases of the neuroplastic changes following the induction of local protein translation in synapse. In addition, mutants with altered CaMKIIR activity show altered synaptic physiology and may be instrumental for the understanding of CaMKIIR-driven synaptic plasticity.31 Finally, as CaMKIIR-dependent phosphorylation in the synapse underlies neuroplasticity, it will be informative to reveal the synaptic phosphoproteome in these mutant mice.

Acknowledgment. This work was funded by the Dutch Genomics Initiative for Horizon breakthrough project to K.W.L., and by the Center of Medical Systems Biology (CMSB) to K.W.L. and A.B.S. We thank Miss Patricia Klemmer for experimental assistance. Supporting Information Available: A table showing the identity and regulation of proteins from hippocampal synaptic membranes of WT and 3′UTR mutant.This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Blitzer, R. D.; Iyengar, R.; Landau, E. M. Biol. Psychiatry 2005, 57, 113-119. (2) Robinson, T. E.; Kolb, B. Eur. J. Neurosci. 1999, 11, 1598-1604. (3) McCullumsmith, R. E.; Clinton, S. M.; Meador-Woodruff, J. H. Int. Rev. Neurobiol. 2004, 59, 19-45. (4) Yamauchi, T. Biol. Pharm. Bull. 2005, 28, 1342-1354. (5) Merrill, M. A.; Chen, Y.; Strack, S.; Hell, J. W. Trends Pharmacol. Sci. 2005, 26, 645-653. (6) Phillips, G. R.; Florens, L.; Tanaka, H.; Khaing, Z. Z.; Fidler, L.; Yates, J. R., 3rd; Colman, D. R. J. Neurosci. Res. 2005, 81, 762775. (7) Witzmann, F. A.; Arnold, R. J.; Bai, F.; Hrncirova, P.; Kimpel, M. W.; Mechref, Y. S.; McBride, W. J.; Novotny, M. V.; Pedrick, N. M.; Ringham, H. N.; Simon, J. R. Proteomics 2005, 5, 2177-2201. (8) Schrimpf, S. P.; Meskenaite, V.; Brunner, E.; Rutishauser, D.; Walther, P.; Eng, J.; Aebersold, R.; Sonderegger, P. Proteomics 2005, 5, 2531-2541. (9) Morciano, M.; Burre, J.; Corvey, C.; Karas, M.; Zimmermann, H.; Volknandt, W. J. Neurochem. 2005. (10) Li, K. W.; Hornshaw, M. P.; van Minnen, J.; Smalla, K. H.; Gundelfinger, E. D.; Smit, A. B. J. Proteome Res. 2005, 4, 725733. (11) DeGiorgis, J. A.; Jaffe, H.; Moreira, J. E.; Carlotti, C. G., Jr.; Leite, J. P.; Pant, H. C.; Dosemeci, A. J. Proteome Res. 2005, 4, 306315.

research articles

Quantitative Synaptic Proteomics (12) Collins, M. O.; Yu, L.; Coba, M. P.; Husi, H.; Campuzano, I.; Blackstock, W. P.; Choudhary, J. S.; Grant, S. G. J. Biol. Chem. 2005, 280, 5972-5982. (13) Yoshimura, Y.; Yamauchi, Y.; Shinkawa, T.; Taoka, M.; Donai, H.; Takahashi, N.; Isobe, T.; Yamauchi, T. J. Neurochem. 2004, 88, 759-768. (14) Li, K. W.; Hornshaw, M. P.; Van Der Schors, R. C.; Watson, R.; Tate, S.; Casetta, B.; Jimenez, C. R.; Gouwenberg, Y.; Gundelfinger, E. D.; Smalla, K. H.; Smit, A. B. J. Biol. Chem. 2004, 279, 9871002. (15) Jordan, B. A.; Fernholz, B. D.; Boussac, M.; Xu, C.; Grigorean, G.; Ziff, E. B.; Neubert, T. A. Mol. Cell. Proteomics 2004, 3, 857-871. (16) Coughenour, H. D.; Spaulding, R. S.; Thompson, C. M. Proteomics 2004, 4, 3141-3155. (17) Stevens, S. M., Jr.; Zharikova, A. D.; Prokai, L. Brain Res. Mol. Brain Res. 2003, 117, 116-128. (18) Dosemeci, A.; Tao-Cheng, J. H.; Vinade, L.; Jaffe, H. Biochem. Biophys. Res. Commun. 2006, 339, 687-694. (19) Collins, M. O.; Husi, H.; Yu, L.; Brandon, J. M.; Anderson, C. N.; Blackstock, W. P.; Choudhary, J. S.; Grant, S. G. J. Neurochem. 2006, 97 Suppl 1, 16-23. (20) Cheng, D.; Hoogenraad, C. C.; Rush, J.; Ramm, E.; Schlager, M. A.; Duong, D. M.; Xu, P.; Wijayawardana, S. R.; Hanfelt, J.; Nakagawa, T.; Sheng, M.; Peng, J. Mol. Cell. Proteomics 2006, 5, 1158-1170. (21) Li, K. W. Anal. Bioanal. Chem. 2007, 387, 25-28. (22) Melanson, J. E.; Avery, S. L.; Pinto, D. M. Proteomics 2006, 6, 4466-4474.

(23) DeSouza, L.; Diehl, G.; Rodrigues, M. J.; Guo, J.; Romaschin, A. D.; Colgan, T. J.; Siu, K. W. J. Proteome Res. 2005, 4, 377-386. (24) Ross, P. L.; Huang, Y. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.; Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin, S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Mol. Cell. Proteomics 2004, 3, 11541169. (25) Miller, S.; Yasuda, M.; Coats, J. K.; Jones, Y.; Martone, M. E.; Mayford, M. Neuron 2002, 36, 507-519. (26) Siripurapu, V.; Meth, J.; Kobayashi, N.; Hamaguchi, M. J. Mol. Biol. 2005, 346, 83-89. (27) Calvano, S. E.; Xiao, W.; Richards, D. R.; Felciano, R. M.; Baker, H. V.; Cho, R. J.; Chen, R. O.; Brownstein, B. H.; Cobb, J. P.; Tschoeke, S. K.; Miller-Graziano, C.; Moldawer, L. L.; Mindrinos, M. N.; Davis, R. W.; Tompkins, R. G.; Lowry, S. F. Nature 2005, 437, 1032-1037. (28) Hu, J.; Qian, J.; Borisov, O.; Pan, S.; Li, Y.; Liu, T.; Deng, L.; Wannemacher, K.; Kurnellas, M.; Patterson, C.; Elkabes, S.; Li, H. Proteomics 2006, 6, 4321-4334. (29) Colbran, R. J.; Brown, A. M. Curr. Opin. Neurobiol. 2004, 14, 318327. (30) Matsuzaki, M.; Honkura, N.; Ellis-Davies, G. C.; Kasai, H. Nature 2004, 429, 761-766. (31) Elgersma, Y.; Sweatt, J. D.; Giese, K. P. J. Neurosci. 2004, 24, 8410-8415.

PR070086W

Journal of Proteome Research • Vol. 6, No. 8, 2007 3133