Analysis of the Dorsal Spinal Cord Synaptic Architecture by Combined

Analysis of the Dorsal Spinal Cord Synaptic Architecture by Combined Proteome Analysis and in Situ Hybridization Mathias Dreger,*,†,# Joanna Mika,‡,| Annette Bieller,‡ Ricarda Jahnel,† Clemens Gillen,§ Martin K. H. Schaefer,‡ Eberhard Weihe,‡ and Ferdinand Hucho† Institute for Chemistry/Biochemistry, Thielallee 63, Freie Universita¨t Berlin, Germany, Department of Molecular Neuroscience, Institute of Anatomy and Cell Biology, Philipps University, Marburg, Germany, Molecular Pharmacology, Gru ¨ nenthal GmbH, Aachen, Germany, University Laboratory of Physiology, University of Oxford, Parks Road, Oxford OX1 3PT, United Kingdom, and Department of Molecular Neuropharmacology, Institute of Pharmacology, Polish Academy of Sciences, 12 Smetna, 31-343 Cracow, Poland Received July 29, 2004

The proteomic analysis of tissue samples is an analytical challenge, because identified gene products not only have to be assigned to subcellular structures, but also to cell subpopulations. We here report a strategy of combined subcellular proteomic profiling and in situ hybridization to assign proteins to subcellular sites in subsets of cells within the dorsal region of rat spinal cord. With a focus on synaptic membranes, which represent a complex membrane protein structure composed of multiple integral membrane proteins and networks of accessory structural proteins, we also compared different twodimensional gel electrophoresis systems for the separation of the proteins. Using MALDI mass spectrometric protein identification based on peptide mass fingerprints, we identified in total 122 different gene products within the different synaptic membrane subfractions. The tissue structure of the dorsal region of the spinal cord is complex, and different layers of neurons can be distinguished neuroanatomically. Proteomic data combined with an in situ hybridization analysis for the detection of mRNA was used to assign selected gene products, namely the optical atrophy protein OPA-1, the presynaptic cytomatrix protein KIAA0378/CAST1, and the uncharacterized coiled-coil-helix-coiled-coilhelix domain containing protein 3 (hypothetical protein FLJ20420), to cell subsets of the dorsal area of the spinal cord. Most striking, KIAA0378/CAST1 mRNA was found only sparsely within the dorsal horn of the spinal cord, but highly abundant within the dorsal root ganglion. This finding, combined with the identification of KIAA0378/CAST1 within the synaptic membrane fraction of the spinal cord at the protein level, are consistent with the reported presynaptic localization of CAST, predominantly within the tissue we investigated primarily attributable to primary afferent sensory neurons. Our approach may be of use in broader studies to characterize the proteomes of neural tissue. Keywords: subcellular proteomics • complex membrane structure • MALDI-TOF-MS • in situ hybridization • nociception • pain • dorsal horn • spinal cord • subcellular fractionation

Introduction The dorsal horn of the spinal cord is a complex tissue region that contains multiple synaptic contacts between primary afferent neurons, interneurons and projection neurons, that convey incoming information from the body periphery to the brain. Of particular biomedical interest is the nociceptionspecific subset of neurons that form synapstic contacts within * To whom correspondence should be addressed. E-mail: [email protected]. † Institute for Chemistry/Biochemistry, Freie Universita¨t Berlin. ‡ Department of Molecular Neuroscience, Institute of Anatomy and Cell Biology, Philipps University. § Molecular Pharmacology, Gru ¨ nenthal GmbH. # University Laboratory of Physiology, University of Oxford.| Department of Molecular Neuropharmacology, Institute of Pharmacology, Polish Academy of Sciences.

238

Journal of Proteome Research 2005, 4, 238-249

Published on Web 02/24/2005

the dorsal horn area of the spinal cord. The cytoarchitecture of the dorsal horn region of the spinal cord is complex, because certain subpopulations of primary afferent nociceptive neurons form synapses with dorsal horn neurons within distinct layers of the dorsal horn, giving rise to the formation of the dorsal horn laminae.1 For the transduction of noxious stimuli, at the molecular level, it appears that there are many signaling modes and pathways that are also used elsewhere in the nervous system. One example is glutamatergic neurotransmission, that plays a role in nociceptive transmission as well as in neurotransmission in many other contexts, such as learning and memory. However, the modes in which these common pathways are finetuned in nociception, have only begun to be understood.2,3 The protein complements of many subcellular structures have been addressed by combined subcellular fractionation and 10.1021/pr049870w CCC: $30.25

 2005 American Chemical Society

Spinal Cord Synaptic Membrane Proteome

comprehensive mass spectrometric protein identification, an approach termed “subcellular proteomics”.4-6 While there have been a number of subcellular proteomic studies on synaptic structures from brain,7-15 there has been no previous effort to analyze synaptic regions of the dorsal region of the spinal cord by a proteomics approach, and there are only rare reports at all on proteomic characterization of spinal cord proteins16,17 to date. From the analytical point of view, two major challenges are to be met in the proteomic analysis of those dorsal horn synaptic regions: First, the method applied has to be appropriate for membrane proteins, and second, the proteins identified should be assigned to particular subcellular sites of particular cell types. In the present study, we aimed to characterize the proteome of the synaptic regions of the dorsal half of the spinal cord. To meet the analytical challenges outlined above, we designed and applied an integrated strategy of subcellular proteomics and in situ hybridization to identify synaptic membrane proteins within the dorsal horn region and assign them to subpopulations of neurons. We enriched for synaptic membrane proteins by subcellular fractionation. Further subfractionation of the synaptic membrane preparation allowed to enrich for protein subpopulations of synaptic membranes. These were complementarily analyzed by conventional two-dimensional gel electrophoresis of isoelectric focusing as the first and SDS-PAGE as the second separation dimension (IEF/SDS-PAGE), and by the membrane protein-compatible 16-benzyl dimethyl hexadecylammoniumchloride (16-BAC)-/SDS-PAGE two-dimensional electrophoresis system. This was followed by in-gel trypsinisation of proteins and MALDI-TOF mass spectrometric protein identification based on peptide mass fingerprints. We identified a set of 122 proteins that represent the abundant proteins of the preparation. To assign several proteins to particular laminae of the dorsal horn region, we performed in situ hybridization experiments to detect those cells that contain the particular transcripts. The strategy we applied represents a valuable aproach to investigate subproteomes within complex tissue regions composed of different cell populations.

Experimental Procedures Chemicals and Antibodies. The monoclonal anti-NMDA receptor subunit R1 antibody and the monoclonal anti-synaptophysin antibody were purchased from Synaptic Systems Inc., Go¨ttingen, Germany. Bovine trypsin (sequencing grade) was purchased from Sigma Aldrich (Deisenhofen, Germany), as well as all other chemicals used for the protein digestion protocol. Synaptic Membrane Preparation. From dorsal halves of spinal cords of adult rats, synaptosomes were prepared as described.18 In brief, the spinal cord dorsal halves were homogenized in homogenization buffer [0.32 M sucrose, 10 mM HEPES/NaOH pH 7.4, Complete protease inhibitor (Merck Biosciences)] in a glass/Teflon douncer at 900 rpm. By lowspeed centrifugation (1000 × g), nuclei and cell debris were removed as pellet P1, whereas synaptosomes under these conditions remained in the supernatant S1. Synaptosomes and other vesicles and organelles were pelletted as P2 fraction by a centrifugation at higher speed (10 000 × g), whereas the supernatant fraction S2 was discarded. The P2 fraction, which contains myelin, light membranes (microsomes), synaptosomes, and mitochondria, was, after resuspension, loaded onto

research articles a discontinuous sucrose gradient with sucrose concentration steps of 0.8/1.0/1.2 M sucrose. These gradients were centrifuged for 2 h at 85 000 × g in an ultracentrifuge. Synaptosomes were recovered from the phase border from 1.0 to 1.2 M sucrose (Figure 1a). In a typical experiment, the preparation started from ∼2 g of tissue (wet weight). From this, ∼2 mg of synaptic membrane protein (as determined by the Bradford method)19 could be obtained, corresponding to ∼2% of the total dorsal spinal cord protein. The crude synaptic membrane preparation was further fractionated according to the scheme described:20 The preparation was fractionated by treatment either with 0.5% (w/v) Triton X-100 fu ¨ r 15 min at 4 °C, to enrich for membrane-associated structural proteins and integral membrane proteins attached to that scaffold, or with 4 M urea/0.1 M Na2CO3 for 5 min. at room temperature, to remove peripheral membrane proteins and enrich for integral membrane proteins in general. The Triton X-100-resistant material was recovered by centrifugation at 13 000 × g in a tabletop centrifuge, the urea/carbonateresistant material was recovered by ultracentrifugation at 50 000 rpm in a Beckman tabletop ultracentrifuge. The synaptic membrane subfractions contained an estimated portion of ∼50% or less of the total amount of synaptic membrane protein prior to extraction. For two-dimensional separation by isoelectric focusing/SDS-PAGE (IEF-/SDS-PAGE), 400 µg of total synaptic membrane protein was extracted. Thus, approximately 200 µg of protein was applied to one IPG strip. For 16-BAC-/ SDS-PAGE, half of the material was used per sample, due to the limited capacity of the gel in the first dimension. Two-Dimensional Gel Electrophoresis. The proteins were resuspended and then precipitated by chloroform/methanol. For isoelectric focusing (IEF), the precipitate was resupended in 7 M urea, 2 M thiourea, 20 mM TrisBase, 0.5% dodecyl maltoside, 0.5% IPG-buffer 3-10 nonlinear (Amersham Biosciences), 40 µg/mL digitonin for 60 min before application to a 7 cm IPG strip 3-10 nonlinear (Amersham Biosciences) and isoelectric focusing using the IPGphor apparatus (Amersham Biosciences). The voltage protocol used was as follows: 1000 Vhr at 500 V, 2000 Vhr at 1000 V, and 16 000 Vhr at 8000 V. As the second separation dimension, 7.5%-15% acrylamide/ bisacrylamide (30:0.8) gradient gels according to Laemmli21 were used. For 16-benzyl dimethyl hexadecylammonium chloride (16-BAC) gel electrophoretic separation, the precipitates were resuspended in sample buffer similar to what has been described,22,23 and separated on 4-10% acrylamide BAC gels in the first and 7.5-15% acrylamide/bisacrylamide (30:0.8) gradient gels according to Laemmli in the second dimension. The minigel gel format was used. All gels were run in at least duplicate. Protein Identification. The protein gels were stained by Coomassie G-250, and the proteins were digested in-gel according to standard procedures.24 The fast evaporation/ nitrocellulose (FENC) matrix preparation with R-cyano-4hydroxy cinnamic acid was used in the initial attempts to obtain peptide mass fingerprints. Further measurements were performed after desalting of the samples via C18-ZipTips (Millipore), and a sandwich matrix preparation was used for these samples as described.25 Mass spectrometric measurements were conducted using a Bruker Reflex MALDI-TOF mass spectrometer equipped with a reflector and pulsed ion extraction. Peptide mass spectra were calibrated internally using the masses of known trypsin autoproteolysis products. Peptide mass fingerprints were matched to the NCBI nonredunJournal of Proteome Research • Vol. 4, No. 2, 2005 239

research articles

Dreger et al.

Figure 1. Preparation scheme for the isolation of synaptosomes from the dorsal half of rat spinal cord and probing of fractions of the preparation for enrichment of synaptic marker proteins synaptophysin and NMDA receptor R1 subunit. Both marker proteins were strongly enriched from the homogenate to the synaptic junctions fraction.

dant database using the search engine ProFound at http:// 129.85.19.192/profound_bin/WebProFound.exe. Mass accuracy was typically e50 ppm after internal calibration with tryptic autoproteolysis products, with additional peptides being matched when the accuracy stringency was lowered to e75 ppm. Proteins were considered positively identified according to the 240

Journal of Proteome Research • Vol. 4, No. 2, 2005

default significance criteria of the search engine. No fixed value for sequence coverage or number of matching peptides was used. Western Blot Analysis. Western blot analysis was performed according to standard procedures. We used an anti-NMDA receptor subunit R1 monoclonal antibody (Synaptic Systems,

research articles


Go¨ttingen, Germany) at a dilution of 1:2000 or an antisynaptophysin monoclonal antibody (Synaptic Systems, Go¨ttingen, Germany) at a dilution of 1:2000 as primary antibodies. An alkaline phosphatase-conjugated anti-mouse-anitibody (Sigma Aldrich, Deisenhofen, Germany), was used as the secondary antibody at a dilution of 1:1000. In Situ Hybridization. By using RT-PCR, cDNA fragments of KIAA0378/CAST (422bp), FLJ20420 (580bp) and OPA1 (405bp) were amplified with the following gene-specific primer pairs, subcloned into the TOPOII vector and used for in vitro transcription as described previously:26 5′-GGCACAAAGT ACTCCTCATACACAC-3′/5′-TATGCACGTTAACAGCGTGTC AAC-3′ (CAST), 5′-GTCTAAGAAGCTCATCGCTCTAGC-3′/5′GTCTTGCCAAACTTGAGCCAGTC-3′ (OPA1), 5′-GTGATCGACCGGATGAAGGAGT-3′/5′-GTTTGGCATGGTTGACACAGTG-3′ (FLJ20420). The probes for CAST and FLJ20420 were labeled by 35S-UTP and for CAST and OPA1 by 35S-UTP and 35S-CTP (Amersham Biosciences, Braunschweig, Germany). After transcription, the probes were subjected to mild alkaline hydrolysis. In situ hybridization was performed as described previously.27 Brains, spinal cords, and DRGs were removed and frozen on dry ice. Then the tissue was cut in to 12 µm thick slices on a cryostat microtome (Leica Microsystems, Nussloch, Germany). Frozen serial sections were thaw-mounted on adhesive slides and stored at -70 °C. Briefly, all of the following steps were performed at room temperature. Tissues were fixed on the slide by immersion in ice-cold 4% phosphate-buffered formaldehyde solution for 1 h and then washed in 10 mM phosphate-buffered saline (PBS), pH 7.4, three times for 10 min each and once in PBS containing 0.4% Triton X-100. After a short rinse in distilled water the sections were washed in 0.1 M triethanolamine pH 8.0 (Sigma, Deisenhofen, Germany) followed by a second wash in 0.1 M triethanolamine pH 8.0 containing acetic anhydride (0.25% vol./vol.) for 10 min at room temperature. After incubation in 2× SSC, sections were dehydrated in graded alcohols (50%, 70%) and air-dried. Radioactive probes were diluted in hybridization buffer (3× SSC, 50 mM NaxHxPO4, 10 mM dithiothreitol, 1× Denhardt’s solution, 0.25 g/L yeast tRNA, 10% dextran sulfate and 50% formamide) to a final concentration of 5 × 104 dpm/µL and 30 to 50 µL hybridization solution was applied to each section, slides were coverslipped and incubated for 14 h at 60 °C. Slides were washed in 2× SSC and 1× SSC for 20 min each followed by incubation in RNase buffer (10 mM Tris/HCl, pH 8.0, 0.5 M NaCl, 1 mM EDTA) containing 1 unit/ml RNase T1 and 20 µg/ mL RNase A (Boehringer, Mannheim, Germany) for 30 min at 37 °C. The slides were washed at room temperature in 1×, 0.5×, and 0.2× SSC for 20 min each, at 60 °C in 0.2× SSC for 60 min and at room temperature in 0.2× SSC and distilled water for 10 min each. The tissue was dehydrated in 50% and 70% 2-propanol. X-ray film exposures were 24 h for all probes. For that purpose, the dry slides are fixed and placed in an X-ray cassette. A sheet of X-ray film (Amersham Hyper-film β-max) is laid over the sections. Photographic prints are made directly from the X-ray film. Exposure time for emulsion were for CAST 21 days, FLJ20420 16 days, and OPA1 19 days. For this type of microscopic analysis, the hybridized sections are coated with nuclear emulsion. Then the slides are stored at 4 °C for various exposure time. All procedures are performed in a darkroom. Then the slides are developed in Kodak developer and fixed in Kodak fixer. The signals (silver grains) on the tissue are visible

under the microscope. However, this exposure technique is not suited for semiquantitative analysis.

Results Characterization of the Synaptic Membrane Preparation from Spinal Cord. The preparation of synaptic structures by subcellular fractionation is well-established for brain tissue, but spinal cord tissue has not been thoroughly characterized in that way. We adapted a preparation protocol that was designed for the isolation of synaptic membranes from brain tissue of rats18 for the preparation of analogous structures from spinal cord dorsal halves from the segments L3-L5. The synaptic membrane preparation starts with the isolation of synaptosomes. One key step in the preparation procedure is a sucrose discontinuous gradient ultracentrifugation (Figure 1a). From the phase border between 1 M sucrose and 1.2 M sucrose, synaptosomes can be collected. Synaptosomes are vesicles that form spontaneously during homogenization of neuronal tissue. They ideally comprise the presynaptic terminals with material of the synaptic cleft and of the postsynaptic membrane still attached.28 By applying an osmotic shock followed by an appropriate centrifugation, the synaptic membranes can be pelletted. These can again be loaded onto a new discontinuous sucrose gradient to allow the recovery of the “synaptic junction” fraction.18 To monitor the enrichment of specific synaptic membrane marker proteins during the subcellular fractionation procedure, we used Western Blot analysis with antibodies specific for the NMDA receptor R1 (NMDAR1) subunit and the synaptic vesicle protein synaptophysin to probe samples from different stages of the preparation (Figure 1b). Although the anti-NMDAR1 immunoreactivity was weak in general, there was a clear enrichment during the preparation of synaptosomes. The anti-NMDAR1 immunoreactivity appeared slightly further enriched in the synaptic membrane fraction, with no further enrichment in the synaptic junction fraction. Anti-synaptophysin immunoreactivity was also clearly enriched during synaptic membrane preparation. However, as the protein is known to be associated with synaptic vesicles, and does not represent a synaptic membrane protein, it appears that either the osmotic shock treatment of synaptosomes may not have led to their complete disruption, or that after disruption, still a significant number of synaptic vesicles may have remained associated with the synaptic membranes. Due to significant material losses in the preparation of synaptic junctions from synaptic membranes from spinal cord dorsal halves, but apparently no significant further enrichment of NMDAR1, the synaptic membrane preparation was used as a starting point for proteome analysis. Subfractionation of Synaptic Membranes and Gel Electrophoretic Separation of the Proteins. We have recently demonstrated that subfractionation of complex membrane structures prior to comprehensive protein identification not only facilitates protein identification by reduction of the sample complexity, but gives also access to biologically meaningful substructures within the subcellular fraction investigated, and thereby enables more precise predictions on characteristics of previously unknown gene products.20 Synaptic membranes represent a complex membrane structure composed of integral membrane proteins of the pre- and postsynaptic membrane, but in addition, composed of protein scaffolds such as the presynaptic web structure/cytomatrix at the active zone9,29 and the postsynaptic density,30 both of which contact and/or anchor integral membrane proteins. Journal of Proteome Research • Vol. 4, No. 2, 2005 241

research articles

Dreger et al.

Figure 2. Subfractionation to isolate subsets of proteins from the synaptic membrane preparation. A, Flowchart for protein identification from subcellular fractions, B representative gel pictures with indicated positions of identified proteins. See Supporting Information figures and tables for details on the assignment of identified proteins.

We expected to be able to remove nonanchored integral membrane proteins by limited detergent extraction of the spinal cord synaptic membrane fraction, thereby getting access to a synaptic scaffold protein fraction with anchored integral membrane proteins. In contrast, by chaotrope-treatment of the synaptic membrane protein fraction, we expected to “strip away” from the membranes the synaptic scaffold proteins, leaving both formerly anchored and nonanchored integral membrane proteins behind for protein identification. While it is apparent that two-dimensional protein separation with isoelectric focusing as the first separation dimension followed by SDS-PAGE resolves protein mixtures by far better than 16BAC-/SDS-PAGE, it has been reported to perform poor with integral membrane proteins,31 whereas 16-BAC-/SDS-PAGE has been shown to be effective in the representation of integral membrane proteins.22,31 To get access to as many proteins as possible, we used both separation systems for the separation of synaptic membrane subfractions. Figure 2a gives an overview 242


of the strategy we used for protein identification. By Coomassie staining, about 80 protein spots could be visualized on 16BAC-/SDS-PAGE gels when the Triton X-100-resistant fraction was separated. About 120-140 spots could be detected by visual inspection on IEF-/SDS-PAGE gels by Coomassie staining after separation of the same synaptic membrane subfraction. When the chaotrope-resistant fraction was separated, 50-60 spots could be visualized by Coomassie on The BAC gels, whereas around 40 spots were visible on IEF-/SDS-PAGE gels (Figure 2b, see Supporting Information for details on protein identification and protein lists). Identification of Protein Subsets Present in Synaptic Membrane Subfractions as Separated by Different Gel Electrophoresis Systems. After excision of gel pieces containing Coomassie-stained proteins followed by tryptic in-gel digestion, the peptide mixtures were subjected to MALDI-TOF mass spectrometric measurements to obtain peptide mass fingerprints for database-aided protein identification. A total of 122


research articles

different proteins were identified. All of these were observable as Coomassie-stained spots on the gels, thus representing the most abundant proteins in the synaptic membrane preparation. Seventy-eight different proteins were identified only on those BAC gels, on which the Triton X-100-resistant fraction was separated. Fourty-six proteins were identified on IEF-/SDSPAGE gels, on which the same fraction, but at the double amount of protein loaded, was separated. In the latter case, the comparably low number of identified proteins as compared to the number of visible spots can be explained by the fact that multiple spots were visible for many proteins. Fifty-one different proteins were identified on BAC-gels, on which the chaotrope-resistant fraction was separated. Twelve different proteins were identified on IEF-/SDS-PAGE gels, on which the chaotrope-resistant fractions was separated. When the results of the protein identification from one particular fraction as separated by both separation systems are combined, 96 different proteins were identified from the Triton-resistant fraction. Sixty-one different proteins were identified in the chaotrope-resistant raction. In total, 72 proteins were identified in only one synaptic membrane subfraction as separated by one particular gel electrophoresis system. Of these, 31 were unique to the BAC gel-separated Triton-resistant subset, while 16 were unique to the IEF-/SDS-PAGE-separated Triton-resistant subset. Twenty-three proteins were identified only from the chaotrope-resistant subset as separated by BAC gel electrophoresis, whereas only two were uniquely identified from that fraction as separated by IEF-/SDS-PAGE. The complete list of identified proteins along with the lists of proteins identified from one particular fraction separated by one particular gel system is given as supplementary data. Grouping of Identified Proteins. The proteins identified were grouped according to nine categories: integral membrane proteins of the plasma membrane, membrane-associates signaling proteins, scaffold proteins (cytoskeleton, presynaptic cytomatrix, postsynaptic scaffold), synaptic vesicle-/exocytosisrelated proteins, intracellular membrane proteins, myelin proteins, cytoplasmic soluble proteins (metabolic enzymes, chaperones), mitochondrial proteins, and previously unknown/ uncharacterized proteins (Figure 3a, see Supporting Information for protein names). To give an idea about the major proteins observed on BAC-/SDS-PAGE gels, Figure 4 shows representative gels on which the most abundant protein spots are annotated. About 10% of the total number of identified proteins represented integral membrane proteins of the plasma membrane. Among these proteins, several were synapse- or at least neuron-specific, such as the AMPA-type ionotropic glutamate receptor or several subunits of the Na+/K+-ATPase. Three abundant cell adhesion molecules were detected (contactin-1, paranodin, neurofascin). One of the largest groups of proteins identified was the scaffold proteins. Some of these were characteristic proteins of the postsynaptic density (PSD95, Citron), while others could be assigned to the neuronspecific cytoskeleton (neurofilaments) or the astrocyte cytoskeleton (glial cell fibrillary acidic protein GFAP), or assigned as ubiquitous cytoskeletal elements (such as actin or tubulin isoforms). One protein of the “scaffold” category was contained in the literature as KIAA0378/CAST1, for “cytomatrix at the active zone- associated protein”,32 which has not been characterized within spinal cord tissue yet. This gene product was chosen for complementary analysis by in situ hybridization. Mitochondrial proteins represented around 25% of all proteins identified. Though mitochondrial proteins are not con-

Figure 3. Classification of identified proteins and comparison of the representation of different protein classes by different separation systems: A., numbers of representatives of different proteins classes grouped according to their known subcellular assignment, PM, plasma membrane; PM assoc., plasma membrane-associated signaling proteins; scaffold/CS, synaptic scaffold/cytoskeletal proteins; SV, synaptic vesicle- and exocytosisrelated proteins; InM, proteins of intracellular membranes, but not mitochondria; My, myelin; Sol/enz, cytosolic proteins and metabolic enzymes; Mito, mitochondrial proteins; un, uncharacterized. B., representation of classes of identified proteins in different subfractions, Tx-res: Triton X-100-resistant fraction, Ch-res: chaotrope-resistant fraction; C, representation of classes of identified proteins by different separation systems, BAC: 16BAC-/SDS-PAGE, IEF: isoelectric focusing/SDS-PAGE.

stituents of synaptic membranes, mitochondria are contained in synaptosomes, and mitochondrial membranes may be in close proximity to synaptic membranes. One of the mitochondrial proteins identified was optical atrophy protein OPA-1, which has been described in the brain,33-36 but not yet within the spinal cord. A number of metabolic enzymes and other soluble proteins were also identified, as well several proteins related to the exocytosis machinery, and several myelin proteins. Among the proteins that could not be assigned to a particular substructure Journal of Proteome Research • Vol. 4, No. 2, 2005 243

research articles

Dreger et al.

Figure 4. Representative BAC-/SDS-PAGE gels with proteins of the Triton X-100 or the chaotrope resistant synaptosome membrane fractions. The most abundant proteins, as identified by MALDI-MS, are indicated.

Figure 5. Identification of the three gene products that were later subjected to analysis at the mRNA level by in situ hybridization. A., positions of the proteins on representative BAC-/SDS-PAGE gels, B., Peptide mass fingerprint spectra used for protein identification, ions of m/z matching with predicted peptides derived from the respective proteins within the error margin set for database search are indicated by asterisks (*).

was the gene product of the rat ortholgue of FLJ20420/coiledcoil-helix-coiled-coil-helix domain containing protein 3 (later on in this paper referred to simply as “FLJ20420”), which represents an up to now uncharacterized hypothetical protein. This gene product was also chosen for further analysis by in situ hybridization for the detection of mRNA within the tissue. 244


See Figure 5 for the identification data for the three gene products probed for at the mRNA level later by in situ hybridization. Representation of Different Protein Classes in Synaptic Membrane Subfractions and Comparison of their Accessibility by Different Separation Techniques. A comparison of the


protein categories represented in the different subfractions of the synaptic membrane revealed the most striking differences in the case of the scaffold proteins, which were present in the Triton X-100-resistant fraction, but, as expected, less abundant in the chaotrope-resistant fraction (Figure 3b). In the latter fraction, other proteins were better accessible to protein chemical identification. (See also Tables S2-S5 for protein subsets in different fractions on different gels.) We observed also a difference in the abundance of proteins categorized as soluble, which were more abundant in the Triton X 100resistant fraction than in the chaotrope-resistant fraction. The comparison of protein categories represented on different gel systems revealed that integral membrane proteins were almost exclusively represented on the BAC gels, but not on the IEF-/ SDS-PAGE gels (Figure 3c), in line with the notion that the BAC gel system displays better compatibility with integral membrane protein samples than the classic IEF-/SDS-PAGE. In Situ Hybridization to Assign Selected Gene Products to Cell Subpopulations: Localization of KIAA0378/CAST1, FLJ20420, and OPA1 mRNAs in Spinal Cord and Dorsal Root Ganglion. While the subcellular proteomics approach is suiteds within the limits of the purity of the subcellular fractionsto assign gene products to subcellular structures, not much information is contained with respect of the assignment of gene products to cell subpopulations. To be able to assign KIAA0378/ CAST1, OPA-1, and FLJ20420, to cell subpopulations, we performed an in situ hybridization analysis to detect mRNAs of these gene products within tissue sections. KIAA0378/CAST1 mRNA expression was high in the majority of dorsal root ganglion cells (Figure 6A), but only relatively weakly detected in the spinal cord and only observed when the more sensitive emulsion technique was used for exposure of the in situ hybridization samples (Figure 6B. Note that Figure 6, dataset 1, shows classic autoradiography exposure, while dataset 2 shows the results of the application of the emulsion technique). This suggests that the CAST1 protein we identified in the proteomic screen from dorsal halves of the spinal cord was predominantly synthesized by primary sensory neurons of dorsal root ganglia that project to the dorsal spinal cord. We detected very high levels of FLJ20420 mRNA in the majority of neurons of the DRG (Figure 6C) and the spinal cord (Figure 6D). The presence of low levels of FLJ20420 mRNA in the white matter of the spinal cord indicates synthesis of FLJ20420 mRNA also by nonneuronal cells, such as glial cells. Overall levels of OPA1 mRNA were low in both DRG (Figure 6E) and spinal cord (Figure 6F). However, in the laminae I-II of superficial dorsal horn, where nociceptive transmission occurs, OPA1 mRNA levels were somewhat higher than in the other regions of the spinal gray matter (Figure 6F). The concentration in neurons to laminae I-II was also confirmed by high-resolution analysis of emulsion dipped slides (see Figure 6, second image set). Localization of KIAA0378/CAST1, FLJ20420, and OPA1 mRNAs in Rat Brain. Not only primary afferent sensory neurons project into the dorsal region of the spinal cord, but also descending neurons project directly or indirectly into the target area that was subject to our investigation.1 Therefore, we also probed rat brain for the distribution of the mRNAs of the three selected gene products. Consistent with the observation from spinal cord and DRG, that the FLJ20420 mRNA is localized to many cell types at a high expression level, we observed high levels of FLJ20420 also in rat brain (Figure 7c,d). KIAA0378/CAST1 mRNA was found to be expressed in a more restricted pattern, mostly in the hippocampus (Figure 7a,b).

research articles

Figure 6. Image set 1: False color presentation of X-ray autoradiograms illustrating the distribution of CAST1 (A, B), FLJ20420 (C, D) and OPA1 (E, F) mRNAs in rat dorsal root ganglion L5 (DRG) and lumbar spinal cord. Note high expression of CAST1 mRNA in the majority of neurons of DRG (A) and the virtual absence of CAST1 mRNA from the spinal cord (B). FLJ20420 mRNA is expressed at extremely high levels in the majority of neurons of DRG (C) and spinal cord (D). Note very low levels of OPA1 mRNA in DRG (E) and spinal cord (F) with some accumulation of enhanced expression in the superficial dorsal horn (lamina I-II). Xray film exposure 24 h. Scale bar ) 330 µm (DRG), 850 µm (spinal cord). Image set 2: another set of tissue slices corresponding to those shown in the image set 1 was probed for the presence of the mRNAs of CAST1 (A,B), and FLJ20420 (C,D), and OPA-1 (E,F), by in situ hybridization, but the more sensitive and higher-resolving emulsion exposure technique (see methods section) was applied. CAST1 mRNA was now sparsely detected also within the spinal cord, but still not within the dorsal horn area. Exposure time for emulsion: CAST 1 (A,B) 21 days; FLJ20420 (C,D) 16 days; OPA-1 (E,F) 19 days. Scale bar: (A,C,E) 200 µm; (B,D,F) 500 µm. Journal of Proteome Research • Vol. 4, No. 2, 2005 245

research articles

Dreger et al.

the spinal cord. The study was confined to a preparation of synaptic membranes and subfractions thereof prepared from the dorsal halves the spinal cord of rats. In addition to the proteomic screening at the subcellular proteomics level, we performed an in situ hybridization analysis of selected gene products, to be able not only to assign gene products to subcellular structures, but also to particular cell populations within a complex tissue. Our approach, in more general terms, could serve as a useful proteome mapping strategy throughout the nervous system and probably for tissue proteomics in general.

Figure 7. False color presentation of X-ray autoradiograms of sagittal (A,C,E) and coronal (B,D,F) sections showing the distribution of CAST1 (A,B), FLJ20420 (C,D) and OPA1 (E,F) mRNAs in rat brain. Note the low expression of CAST1 mRNA restricted to specific brain regions. FLJ20420 mRNA is abundantly expressed in neurons throughout the brain gray matter with some accumulation in selected regions and nuclei. Note nonneuronal expression in the choroid plexus. OPA1 mRNA is expressed at extremely high levels in the cerebellum and at low to moderate levels in restricted brain regions. Olfactory bulb (OLB), cortex (CTX), inferior colliculus (IC), retrospelnial granular Cx (RSG), hippocampus (Hi), caudate putamen (CPu), ventral pallidum (VP), thalamus (Th), hypothalamus (Hy), substantia nigra (SN), medial amygdala nu (Me), ventro medial hypothalamus nu (VMH), brain stem nuclei [pontine nuclei (Pn), 7 facial nu (7), lateral reticular nuclei (LRt)], zona incerta (ZI), cerebellum (CBL), choroid plexus (ChP). X-ray film exposure 24 h. Scale bar ) 3 mm.

OPA-1 mRNA was observed in a restricted manner as well, predominantly in the olfactory bulb and in the cerebellum (Figure 7e,f). The in situ hybridization data are summarized in Table 1. Combination of Proteomic and in Situ Hybridization Data. When proteomic and in situ hybridization data are combined our results suggest that FLJ20420 is an abundant gene product that is not restricted to subsets of cells within the tissue we investigated. It is probaly part of or attached to cytoskeletal or scaffold structures, as they occur at synapses, but it might be localized at other cellular sites as well. OPA-1 appears to be restricted to subsets of cells, not only within the brain, but also within the spinal cord within areas involved in nociception. Because OPA-1 is known to be localized to mitochondria, and was coenriched as part of mitochondrial membranes that occur in the synaptic membrane preparation. The presynaptic cytomatrix protein KIAA0378/CAST1 was detected within the spinal cord at the protein level, but the mRNA was most abundantly observed in the DRG, and at restricted sites within the brain, whereas only low amounts of mRNA were detected in the spinal cord. This implies that the protein is part of the presynaptic cytomatrix of sensory afferent neurons that project into the dorsal horn area of the spinal cord.

Discussion We here present the results of the, up to our best knowledge, first proteomic study on synaptic regions of the dorsal part of 246


One hundred twenty-two different proteins were detected in total after gel electrophoretic protein separation, in-gel protein digestion, and MALDI-TOF-MS-based peptide mass fingerprinting. Our fractionation and gel-based separation strategy gave access to different subsets of synaptic membrane proteins. This is because particular types of proteins were enriched in subfractions (e.g., scaffold proteins in the Triton X 100-resistant fraction). The alternative extraction of the synaptic membranes thus led to a functionally meaningful subfractionation, because whole classes of proteins were present were largely absent in different subfractions. Moreover, the overall sample complexity was reduced by this subfractionation, which is favorable for MALDI-MS and peptide mass fingerprint-based protein identification. Striking differences were observed in the categories of proteins represented by different separation methods: Integral membrane proteins were almost exclusively represented on BAC gels, whereas they were not detected on an IEF-/SDS-PAGE gel when the same sample, even at the double amount of total protein applied to the gel. This is in line with previous observations reviewed recently.31 However, several proteins were nevertheless uniquely detected on IEF-/ SDS-PAGE gels, most likely to the much higher resolution of this technique as compared to the BAC gel system. It should be noted that protein identification from complex mixtures, at the present state of the art, is more efficient when liquid chromtography/mass spectrometry is performed.37 This is in line with recent proteomic studies on synaptic proteins of the brain, in which more proteins have been identified than in our present study.13-15 Which synaptic membrane proteins have been accessible, under these circumstances, by our proteome analysis strategy? With the respect to the number of identified proteins and under consideration of the limited power of our approach for identification of multiple proteins contained in one protein spot, we have detected the most abundant proteins in our preparation. The identification of several known components of the postsynaptic density structure known from brain synapses30 clearly supported findings from the Western blot analysis that synaptic structures from the spinal cord were successfully enriched. We identified, e.g., the proteins PSD95, citron, isoforms of the Ca2+/calmodulin-dependent protein kinase, and the GluR2/3 subunit of the ionotropic glutamate receptor of the AMPA-type. However, one would expect many more proteins derived from the postsynaptic density structure, e.g., if one considers the complex organization of the NMDA receptor complex as determined by the affinity purification approach of Husi et al.,8 and as has been observed in a recent study on a postsynaptic density preparation from rat brain prepared according to a protocol very similar to what was used in our study.13 Due to a lower degree of enrichment, less postsynaptic density proteins have been identified as compared to the studies by Walikonis et al.,7 or Satoh et al.,10 who

research articles


Table 1. Distribution of KIAA0378/CAST1, FLJ20420, and OPA1 mRNAs in Brain, Spinal Cord, and Dorsal Root Ganglionsin Situ Hybridization

KIAA0378/CAST1

brain

spinal cord

DRG

olfactory nuclei, cortex, inferior colliculus, hippocampus [CA1-3, dentate gyrus], substantia nigra, brain stem nuclei [pontine nuclei]

sparse

abundant

BAB21974/FLJ20420 olfactory nu, cortex, cingulate cortex, inferior colliculus, extremely abundant extremely abundant hippocampus [CA1-3, dentate gyrus], caudate putamen, ventral pallidum, thalamus [mediodorsal, central medial and centrolateral thalamus nuclei], hypothalamus [arcuate and dorsomedial hypothalamic nuclei], substantia nigra, chorvid plexus, zona incerta, brain stem nuclei [pontine nuclei, 7 facial nu, lateral reticular nuclei], cerebellum KIAA0567/OPA1

olfactory nuclei, cortex, inferior colliculus hippocampus [CA1-3, dentate gyrus], caudate putamen, hypothalamus, medial amygdala nu, ventro medial hypothalamus nu, cerebellum

investigated postsynaptic density fractions prepared from rat brain. However, the postsynaptic density fraction can be prepared from brain with much higher yield than from spinal cord (data not shown), and it is likely that this preparation is selective only for glutamatergic synapses, because other types of synapses lack prominent postsynaptic densities. Interestingly, a set of three cell adhesion proteins (paranodin, neurofascin, and contactin-1) has been assigned to axo-glial junctions adjacent to the nodes of ranvier in the axon membrane recently.38 This structure likely represents an abundant structure in the tissue region investigated, but is not related to synapses. However, at least some of these landmark proteins have also been identified in very recent postsynaptic density proteomics studies.13-15 It remains an open question whether these proteins are actually localized at synapses. An alternative interpretation is that a structure called “paranodal junction”, a cell surface area comprising the contact sites of axons and glia cells at the node of Ranvier,39 is co-enriched with synaptosomes in the sucrose gradient used for subcellular fractionation of membrane vesicles after homogenization of neural tissue. This would explain the presence of so many myelin proteins not only in our study, but also in the proteomics studies on brain postsynaptic density preparations cited here.13-15 Assuming that there is indeed a coexistence and coenrichment of different neuronal and neuro-glial cell surface structures, it could be interesting to attempt to separate true synaptosomes from this unanticipated co-enriched structure that seems to be prevalent in standard synaptosome preparations. The Na+/K+-ATPase subunits are also among the most abundant neuronal proteins. The protein dipeptidyl aminopeptidase-like protein 2 (DPPX) has been recently described as associated with A-type voltage-gated potassium channels,40 and thus probably also represents an abundant neuronal cell surface protein. Given the assumption that our analysis is likely to be restricted to the abundant proteins from the synaptosome membrane preparation, it is noteworthy that PGRL, an immunoglobulin superfamily integral membrane protein that has only been described in the immune system,41 was identified in our study as one of the abundant membrane proteins. Our data suggest that this protein is associated with an abundant neuronal structure. Taken together, though our study has been obvously restricted to the “higher abundance” level of proteins within synaptic membranes, it appears occasionally surprising and informative which proteins show actually up with high abundance. It is a major task in tissue proteomics to consider not only the subcellular localization of a protein identified, but also to

sparse, selective for laminae I-II

sparse

assign it to subpopulations of cell which make up the tissue. This requirement is apparent given the highly organized tissue structure of, e.g., the gray matter of the spinal cord.1 We therefore performed in situ hybridization experiments for the detection of the mRNAs of three selected gene products, the presynaptic protein KIAA0378/CAST1, the hypothetical protein FLJ20420, and the mitochondrial protein OPA-1, both in the tissue area subjected to subcellular proteomic analysis, but also in tissue areas which harbor cell bodies of neurons that project into the target area in which the gene products were identified at the protein level. Both OPA-1 and FLJ20420 mRNAs were detected in the spinal cord as well as in the dorsal root ganglia and in the brain. Whereas FLJ20420 appears to be a very abundant, though still uncharacterized, protein, OPA1 appeared less abundant at the mRNA level. The detection of OPA-1 mRNA within few restricted areas within the rat brain is consistent with previous findings by Misaka et al.35 There is no reference data from the spinal cord. Interestingly, OPA-1 mRNA was observed as enriched within those laminae (I and II) that are known to be involved in the processing of nociceptive information. The protein KIAA0378/CAST1 was described as a protein specifically localized to mainly presynaptic structures of the brain, and was reported to interact with other known components of the presynaptic cytomatrix.32,42,43 These observations suggested a localization of the protein at all presynaptic sites. We detected KIAA0378/CAST1 at the protein level also in the spinal cord. Our in situ hybridization experiments with rat brain revealed a distribution of KIAA0378/CAST1 mRNA confined to particular brain regions, such as the hippocampus. This distribution was somewhat more restricted than that described by Ohtsuka et al.32 Moreover, KIAA0378/CAST1 mRNA was only sparsely detected within the spinal cord, but was highly abundant within the dorsal root ganglion and within distinct brain regions. In this case, our approach of combined in situ hybridization and subcellular proteomics strongly suggests the presynaptic localization of KIAA0378/CAST1 within the dorsal horn of the spinal cord at presynaptic sites of primary afferent sensory neurons. In summary, our study represents a first proteomics study on synaptic structures of the spinal cord. Although only abundant proteins were detected, at this level of sensitivity previously uncharacterized proteins can be detected. Our study provides a number of notions about analytical aspects that also apply to other proteomics studies on complex tissues. We demonstrate that combining subcellular proteomics with in situ hybridization for the detection of particular mRNA species in Journal of Proteome Research • Vol. 4, No. 2, 2005 247

research articles cell subpopulations may provide important insights into characteristics of gene products under investigation, and, if conducted at a larger scale, should prove to be a key approach to describe the molecular architecture of neural tissue. With respect to the aim of introducing proteomics approaches into the area of pain research, our approach deserves further refinement. Most current findings on changes in gene product abundance in paradigms of pain are derived from comparative immunohistochemistry, comparative in situ hybridization, and cDNA microarray analysis (see refs 2,3 for reviews). Our approach as reported here is not likely to permit direct tracking of protein abundance changes in experimental pain paradigms through “seeing” or identifying difference spots on gels, since (a) the high-resolution 2D IEF-/SDS-PAGE gels seems to under-represent key classes of molecules involved in pain transmission, and (b) the 2D BAC-/SDS-PAGE is deficient in resolution. The problem of identification of only a comparatively low number of proteins can certainly be overcome by appropriate shotgun proteomics approaches, that might involve the MudPIT technology.44 It would be a reasonable approach for large-scale proteome research projects such as the human brain proteome project to adopt our approach, with improved protein identification technology, to map dorsal spinal cord proteins along with large-scale in situ hybridization for mRNA detection to construct a more comprehensive proteome map of this tissue area. This reference data could be used for screens in experimental pain paradigms. In terms of quantitative proteomics for the direct detection of protein abundance changes, the technology that appears to us most likely to have the potential for screening for changes in pain paradigms is the ICAT technology.45,46 One has to be aware, though, that the proteome changes that appear in pain paradigms can probably only tackled at the level of subcellular proteomics (similar to our fractionation scheme). In addition, the changes are also likely to involve only small subpopulations of cells within the tissue, and might be restricted to only a few segments of the spinal cord, sometimes even only to the half of the spinal cord segment ipsilateral to, e.g., an experimental nerve lesion, and does not occur on the contralateral side (see ref 1). Thus, proteomics of pain, though highly deserved, will be a substantial technological and strategical challenge.

Acknowledgment. This work was supported by the German Ministry for Education and Research (BMBF), the Deutsche Forschungsgemeinschaft (SFB515), and the Fonds of the Chemical Industry in Germany. M.D. is now supported by the Wellcome Trust. Supporting Information Available: Ch-resistant pellet, synaptic membrane dorsal half, 2DBAC-gel separation (Table S1), Ch-resistant pellet, synaptic membrane dorsal half, 2DIEF-gel separation (Table S2), Tx-resistant pellet, synaptic membrane dorsal half, 2D gel separation (Table S3), Txresistant pellet, synaptic membrane dorsal half, 2DBAC-gel separation (Table S4), Ch BAC-annot, Ch-IEF-annot, Tx-BACannot, and Tx-IEF-annot (Figures S1-S4). List of identified proteins and classification according to their subcellular localization (Table S5). This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Willis, W. D.; Westlund, K. N. Neuroanatomy of the pain system and of the pathways that modulate pain. J. Clin. Neurophysiol. 1997, 14, 2-31.

248


Dreger et al. (2) Woolf, C. J.; Salter, M. W. Neuronal plasticity: increasing the gain in pain. Science 2000, 288, 1765-1769. (3) Julius, D. Basbaum, A. I. Molecular mechanisms of nociception. Nature 2001, 413, 203-210. (4) Dreger, M. Proteome analysis at the level of subcellular structures. Eur. J. Biochem. 2003a, 270, 589-599. (5) Taylor, S. W.; Fahy, E.; Ghosh, S. S. Global organellar proteomics. Trends Biotechnol. 2003, 21, 82-88. (6) Dreger, M. Subcellular proteomics. Mass Spectrom. Rev. 2003b, 22, 27-56. (7) Walikonis, R. S.; Jensen, O. N.; Mann, M.; Provance, D. W., Jr.; Mercer, J. A.; Kennedy, M. B. Identification of proteins in the postsynaptic density fraction by mass spectrometry. J. Neurosci. 2000, 20, 4069-4080. (8) Husi, H.; Ward, M. A.; Choudhary, J. S.; Blackstock, W. P.; Grant, S. G. Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat. Neurosci. 2000, 3, 661-669. (9) Phillips, G. R.; Huang, J. K.; Wang, Y.; Tanaka, H.; Shapiro, L.; Zhang, W.; Shan, W. S.; Arndt, K.; Frank, M.; Gordon, R. E.; Gawinowicz, M. A.; Zhao, Y.; Colman, D. R. The presynaptic particle web: ultrastructure, composition, dissolution, and reconstitution. Neuron 2001, 32, 63-77. (10) Satoh, K.; Takeuchi, M.; Oda, Y.; Deguchi-Tawarada, M.; Sakamoto, Y.; Matsubara, K.; Nagasu, T.; Takai, Y. Identification of activity-regulated proteins in the postsynaptic density fraction. Genes Cells 2002, 7, 187-197. (11) Yoshimura, Y.; Shinkawa, T.; Taoka, M.; Kobayashi, K.; Isobe, T.; Yamauchi, T. Identification of protein substrates of Ca(2+)/ calmodulin-dependent protein kinase II in the postsynaptic density by protein sequencing and mass spectrometry. Biochem. Biophys. Res. Commun. 2002, 290, 948-954. (12) Stevens, S. M., Jr.; Zharikova, A. D.; Prokai, L. Proteomic analysis of the synaptic plasma membrane fraction isolated from rat forebrain. Brain Res. Mol. Brain Res. 2003, 117, 116-128. (13) 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. Proteomics analysis of rat brain postsynaptic density. Implications of the diverse protein functional groups for the integration of synaptic physiology. J. Biol. Chem. 2004, 279, 987-1002. (14) Peng, J.; Kim, M. J.; Cheng, D.; Duong, D. M.; Gygi, S. P.; Sheng, M. Semiquantitative proteomic analysis of rat forebrain postsynaptic density fractions by mass spectrometry. J. Biol. Chem. 2004. (15) Yoshimura, Y.; Yamauchi, Y.; Shinkawa, T.; Taoka, M.; Donai, H.; Takahashi, N.; Isobe, T.; Yamauchi, T. Molecular constituents of the postsynaptic density fraction revealed by proteomic analysis using multidimensional liquid chromatography-tandem mass spectrometry. J. Neurochem. 2004, 88, 759-768. (16) Lee, S. C.; Yoon, T. G.; Yoo, Y. I.; Bang, Y. J.; Kim, H. Y.; Jeoung, D. I.; Kim, H. J. Analysis of spinal cord proteome in the rats with mechanical allodynia after the spinal nerve injury. Biotechnol. Lett. 2003, 25, 2071-2078. (17) Tsugita, A.; Kawakami, T.; Uchida, T.; Sakai, T.; Kamo, M.; Matsui, T.; Watanabe, Y.; Morimasa, T.; Hosokawa, K.; Toda, T. Proteome analysis of mouse brain: two-dimensional electrophoresis profiles of tissue proteins during the course of aging. Electrophoresis 2000, 21, 1853-1871. (18) tom Dieck, S.; Sanmarti-Vila, L.; Langnaese, K.; Richter, K.; Kindler, S.; Soyke, A.; Wex, H.; Smalla, K. H.; Kampf, U.; Franzer, J. T.; Stumm, M.; Garner, C. C.; Gundelfinger, E. D. Bassoon, a novel zinc-finger CAG/glutamine-repeat protein selectively localized at the active zone of presynaptic nerve terminals. J. Cell Biol. 1998, 142, 499-509. (19) Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976, 72, 248-254. (20) Dreger, M.; Bengtsson, L.; Schoneberg, T.; Otto, H.; Hucho, F. Nuclear envelope proteomics: novel integral membrane proteins of the inner nuclear membrane. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 11943-11948. (21) Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970, 227, 680685. (22) Hartinger, J.; Stenius, K.; Hogemann, D.; Jahn, R. 16-BAC/SDSPAGE: a two-dimensional gel electrophoresis system suitable for the separation of integral membrane proteins. Anal. Biochem. 1996, 240, 126-133. (23) Dreger, M.; Otto, H.; Neubauer, G.; Mann, M.; Hucho, F. Identification of phosphorylation sites in native lamina-associated polypeptide 2 beta. Biochemistry 1999, 38, 9426-9434.

research articles

Spinal Cord Synaptic Membrane Proteome (24) Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal. Chem. 1996, 68, 850-858. (25) Watty, A.; Neubauer, G.; Dreger, M.; Zimmer, M.; Wilm, M.; Burden, S. J. 2000. The in vitro and in vivo phosphotyrosine map of activated MuSK. Proc. Natl. Acad. Sci. U.S.A. 97, 4585-4590. (26) Schafer, M. K.; Bette, M.; Romeo, H.; Schwaeble, W.; Weihe, E. Localization of kappa-opioid receptor mRNA in neuronal subpopulations of rat sensory ganglia and spinal cord. Neurosci. Lett. 1994, 167, 137-140. (27) Schafer, M. K. Day, R. In situ hybridization techniques to map processing enzymes. Methods Neurosci. Ref Type: Serial (Book, Monograph) 1995, 23, 16-43. (28) Gray, E. G. Whittaker, V. P. The isolation of nerve endings from brain: An electron-microscopic study of cell fragments derived by homogenization and centrifugation. J. Anat. 1962, 96, 79-88. (29) Dresbach, T.; Qualmann, B.; Kessels, M. M.; Garner, C. C.; Gundelfinger, E. D. The presynaptic cytomatrix of brain synapses. Cell Mol. Life Sci. 2001, 58, 94-116. (30) Kennedy, M. B. Signal-processing machines at the postsynaptic density. Science 2000, 290, 750-754. (31) Santoni, V.; Molloy, M.; Rabilloud, T. Membrane proteins and proteomics: un amour impossible? Electrophoresis 2000, 21, 1054-1070. (32) Ohtsuka, T.; Takao-Rikitsu, E.; Inoue, E.; Inoue, M.; Takeuchi, M.; Matsubara, K.; Deguchi-Tawarada, M.; Satoh, K.; Morimoto, K.; Nakanishi, H.; Takai, Y. Cast: a novel protein of the cytomatrix at the active zone of synapses that forms a ternary complex with RIM1 and munc13-1. J. Cell Biol. 2002, 158, 577-590. (33) Alexander, C.; Votruba, M.; Pesch, U. E.; Thiselton, D. L.; Mayer, S.; Moore, A.; Rodriguez, M.; Kellner, U.; Leo-Kottler, B.; Auburger, G.; Bhattacharya, S. S.; Wissinger, B. OPA1, encoding a dynaminrelated GTPase, is mutated in autosomal dominant optic atrophy linked to chromosome 3q28. Nat. Genet. 2000, 26, 211-215. (34) Delettre, C.; Lenaers, G.; Griffoin, J. M.; Gigarel, N.; Lorenzo, C.; Belenguer, P.; Pelloquin, L.; Grosgeorge, J.; Turc-Carel, C.; Perret, E.; Astarie-Dequeker, C.; Lasquellec, L.; Arnaud, B.; Ducommun, B.; Kaplan, J.; Hamel, C. P. Nuclear gene OPA1, encoding a mitochondrial dynamin-related protein, is mutated in dominant optic atrophy. Nat. Genet. 2000, 26, 207-210. (35) Misaka, T.; Miyashita, T.; Kubo, Y. Primary structure of a dynamin-related mouse mitochondrial GTPase and its distribution in brain, subcellular localization, and effect on mitochondrial morphology. J. Biol. Chem. 2002, 277, 15834-15842.

(36) Mootha, V. K.; Bunkenborg, J.; Olsen, J. V.; Hjerrild, M.; Wisniewski, J. R.; Stahl, E.; Bolouri, M. S.; Ray, H. N.; Sihag, S.; Kamal, M.; Patterson, N.; Lander, E. S.; Mann, M. Integrated analysis of protein composition, tissue diversity, and gene regulation in mouse mitochondria. Cell 2003, 115, 629-640. (37) Aebersold, R.; Mann, M. Mass spectrometry-based proteomics. Nature 2003, 422, 198-207. (38) Arroyo, E. J.; Xu, T.; Grinspan, J.; Lambert, S.; Levinson, S. R.; Brophy, P. J.; Peles, E.; Scherer, S. S. Genetic dysmyelination alters the molecular architecture of the nodal region. J. Neurosci. 2002, 22, 1726-1737. (39) Bhat, M. A. Molecular organization of axo-glial junctions. Curr. Opin. Neurobiol. 2003, 13, 552-559. (40) Nadal, M. S.; Ozaita, A.; Amarillo, Y.; de Miera, E. V.; Ma, Y.; Mo, W.; Goldberg, E. M.; Misumi, Y.; Ikehara, Y.; Neubert, T. A.; Rudy, B. The CD26-related dipeptidyl aminopeptidase-like protein DPPX is a critical component of neuronal A-type K+ channels. Neuron 2003, 37, 449-461. (41) Clark, K. L.; Zeng, Z.; Langford, A. L.; Bowen, S. M.; Todd, S. C. PGRL is a major CD81-associated protein on lymphocytes and distinguishes a new family of cell surface proteins. J. Immunol. 2001, 167, 5115-5121. (42) Wang, Y.; Liu, X.; Biederer, T.; Sudhof, T. C. A family of RIMbinding proteins regulated by alternative splicing: Implications for the genesis of synaptic active zones. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 14464-14469. (43) Ko, J.; Na, M.; Kim, S.; Lee, J. R.; Kim, E. Interaction of the ERC family of RIM-binding proteins with the liprin-alpha family of multidomain proteins. J. Biol. Chem. 2003, 278, 42377-42385. (44) Washburn, M. P.; Wolters, D.; Yates, J. R. 3rd. Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 2001, 19, 242-247. (45) Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; and Aebersold, R. Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol. 1999, 17, 994999. (46) Hansen, K. C.; Schmitt-Ulms, G.; chalkley, R. J.; Hirsch, J.; Baldwin, M. A.; Burlingame, A. L. Mass spectrometric analysis of protein mixtures at low levels using cleavable 13C-isotope-coded affinity tag and multidimensional chromatography. Mol. Cell Proteomics 2003, 2, 299-314.

PR049870W

Journal of Proteome Research • Vol. 4, No. 2, 2005 249

Analysis of the Dorsal Spinal Cord Synaptic Architecture by Combined

Recommend Documents