Proteomic Analysis of Mouse Brain Microsomes: Identification and Bioinformatic Characterization of Endoplasmic Reticulum Proteins in the Mammalian Central Nervous System Stanley M. Stevens, Jr.,† R. Scott Duncan,‡ Peter Koulen,‡ and Laszlo Prokai*,† Department of Molecular Biology and Immunology and Department of Pharmacology and Neuroscience, University of North Texas Health Science Center, 3500 Camp Bowie Boulevard, Fort Worth, Texas 76107 Received September 30, 2007
The endoplasmic reticulum (ER) is the main source for the storage and release of intracellular calcium in neurons and, thus, contributes to the functionality of a diverse set of pathways that control critical aspects of central nervous system function including but not limited to gene expression, neurotransmission, learning, and memory. ER-derived proteins obtained after subcellular fractionation of mouse brain homogenate were digested with trypsin and the corresponding peptides fractionated by strong cation exchange chromatography followed by LC-MS/MS analysis on a hybrid linear ion trap-Fourier transform ion cyclotron resonance (FTICR) mass spectrometer. A comprehensive catalogue representing 1914 proteins was generated from this particular proteomic analysis using identification criteria that corresponded to a false positive identification rate of 0.4%. Various molecular functions and biological processes relevant to the ER were identified upon gene ontology (GO)-based analysis including pathways associated with molecular transport, protein trafficking and localization, and cell signaling. Comparison of the 2D-LC-MS/MS results with those obtained from shotgun LC-MS/MS analyses demonstrated that most molecular functions and biological processes were represented via GO analysis using either methodology. Results from this comparison as well as a focused investigation into components of calcium-mediated signaling in the mouse brain ER are also presented. Keywords: brain microsomes • endoplasmic reticulum • calcium signaling • proteomics • subcellular fractionation • Fourier transform ion cyclotron resonance mass spectrometry
Introduction Cellular complexity and dynamic range of protein expression represent complicated analytical challenges in proteomicsbased research.1 Although significant improvements in mass spectrometric instrumentation have facilitated analysis of complex protein mixtures such as those obtained from mammalian tissue samples, subcellular fractionation in addition to multidimensional protein and peptide separation strategies remain important for the identification and quantification of low-abundance proteins as well as protein posttranslational modifications.2–5 In particular, the combination of orthogonal chromatographic techniques such as strong cation exchange (SCX) and reversed-phase (RP) HPLC5 or 1D-SDS-PAGE protein fractionation in conjunction with high-resolution tandem mass spectrometry (MS/MS)6 collectively provide a powerful foundation for proteome-wide analyses. These methodologies have been particularly effective for the detailed proteomic characterization of a variety of organ tissues from rodent-based experimental models7–11 and consequently have provided * To whom correspondence should be addressed. Tel.: (817) 735-2206. Fax: (817) 735-2118. E-mail:
[email protected]. † Department of Molecular Biology and Immunology. ‡ Department of Pharmacology and Neuroscience.
1046 Journal of Proteome Research 2008, 7, 1046–1054 Published on Web 02/14/2008
further insight into numerous disease pathologies as well as the development of potential treatment strategies for those ailments. Techniques for proteomic analysis with focus on the central nervous system (CNS) have significantly advanced our understanding of the complex cellular and biochemical processes associated with the mammalian brain.12 The first step in minimization of this complexity and the problems related with the large dynamic range of protein expression for subsequent mass spectrometry analysis have typically involved hypothesisdriven selection and subsequent isolation of specific regions and/or organelle structures from the brain. Several “subproteome” studies have been reported on region-specific brain preparations13–15 including more focused organelle-based proteomic analyses of neuronal mitochondria,7,11,16 plasma membrane17–20 (which also includes detailed analysis of isolated presynaptic21,22 and postsynaptic density structures23–26), microsomes,7,16 and nuclear fractions.7 Using standard density gradient centrifugation methods, subcellular fractions from rodent brain containing enriched organelle structures can be obtained, and consequently, this technique has presented a means for targeted analysis of cellular processes governed by specific neuronal cell constituents. The endoplasmic reticulum (ER), in particular, is an intracellular structure that mediates numerous cellular activities.27,28 10.1021/pr7006279 CCC: $40.75
2008 American Chemical Society
Proteomic Analysis of Mouse Brain Microsomes
research articles
Figure 1. Chromatographic separation of ER tryptic digest. (a) Strong cation exchange (SCX) fractionation of ER tryptic digest as monitored at 280 nm UV absorbance. (b) Reversed-phase (RP) HPLC-MS/MS analysis of SCX fraction 25. (c) Shotgun RP-HPLC-ESI-MS/MS analysis of the ER tryptic digest using a 1 h linear gradient.
Figure 2. Overrepresented subcellular localizations of ER proteins identified from 2D-LC tandem MS analysis (p < 0.05).
However, very few detailed proteomic studies have been carried out on this organelle isolated from mammalian brain tissue limiting, thus, information on proteins involved in ER-mediated neuronal function in vivo. The secretory pathway, for example, by which certain proteins are exported and secreted out of the cell after synthesis in the ribosomes, originates in the rough ER. Proteins designated for the secretory pathway proceed through various cellular compartments and are packaged in vesicles that ultimately fuse to the plasma membrane for cell secretion. Proteins intended for cell secretion and those destined for localization at the plasma membrane require appropriate folding as well as other factors of transport regulation in the ER. In addition to protein folding and
transport, the ER engages in a variety of other cellular processes that include calcium storage and release, synthesis and storage of certain biomolecules, as well as some aspects of cellular detoxification. Dysfunction of the ER has been implicated in certain neurological diseases;29,30 thus, thorough proteomic analysis of brain-derived ER should further improve our understanding of the neuronal functions driven by this organelle and the corresponding pathways potentially involved in those diseases. In addition, the ER represents the major source of intracellular calcium release in neurons and participates in numerous signaling pathways that are dependent on the intracellular calcium ion concentration, such as gene expression and neurotransmission.27,31–33 Journal of Proteome Research • Vol. 7, No. 3, 2008 1047
research articles
Stevens et al.
Table 1. Overrepresented Biological Processes Associated with the Mouse Brain ER Proteome as Determined by Gene Ontology (GO) Analysis using BINGO v2.0 GO-ID
description
51234 16043
establishment of localization cellular component organization and biogenesis cofactor metabolism generation of precursor metabolites and energy oxidative phosphorylation cell-cell signaling cofactor catabolism biosynthesis nucleotide metabolism
51186 6091 6119 7267 51187 9058 9117
corrected p-value
no. of proteinsa
6.4805E-48 6.6874E-26
485 306
1.7806E-17 1.8939E-16
61 147
1.3202E-15 1.4186E-13 8.0914E-10 2.3251E-10 3.8425E-9
35 71 15 231 52
a Out of 1398 annotated ER proteins using GO terms for biological processes with GOA mouse v38.0 (26 030 proteins from the IPI database were annotated by a biological process).
We present in this report a detailed catalogue of the ER proteome obtained from mouse brain using multidimensional separation techniques in conjunction with high-resolution tandem mass spectrometry. Our findings represent the most comprehensive analysis of brain ER proteins to date that includes characterization of biological processes associated with a majority of the 1914 proteins identified. Comparison of shotgun-based LC-MS/MS analysis with 2D-LC-MS/MS showed significant representation of relevant biological processes using either method, thereby indicating that the shotgun approach may be adequate for certain “screening-based” proteomic applications. Pathways identified and validated by MS/MS are also presented with a focus on calcium signaling mediated by ER-resident transmembrane proteins.
Experimental Methods Chemicals and Reagents. All chemicals and reagents were purchased from Sigma Aldrich (St. Louis, MO) unless otherwise specified. Isolation of Mouse Brain Microsomes. Endoplasmic reticulum microsomes were prepared as described previously.34–37 Brain tissue was isolated from 8-week-old male C57BL/6 mice (The Jackson Laboratory, Bar Harbor, Maine) that had been kept on standard rodent chow and environmental conditions on a 12 h day/night cycle in an AALAC approved facility. All experiments were performed in compliance with the guidelines for the welfare of experimental animals issued by the National Institutes of Health and in accordance with institutional guidelines. Briefly, brains from adult mice were removed, placed in ice-cold homogenization buffer (250 mM sucrose, 5 mM HEPES-KOH, pH 7.4, 1 mM EGTA, 1 mM DTT and protease inhibitors), and mechanically homogenized on ice using a Dounce homogenizer. After homogenization, the brain lysate was subjected to differential centrifugation. The tissue homogenate was centrifuged at 1000g for 10 min. The supernatant was then collected and centrifuged at 8000g for 10 min. The resulting supernatant was collected and centrifuged again at 100 000g for 75 min. The resulting pellet (containing the ER microsomes) was resuspended in storage buffer (250 mM sucrose, 5 mM HEPES-KOH, pH 7.4, 1 mM DTT and protease inhibitors), carefully rehomogenized and stored in aliquots at -80 °C for later use. 1048
Journal of Proteome Research • Vol. 7, No. 3, 2008
Preparation of ER Protein Digests. Seven 25 µL aliquots of the ER microsome fraction were used for subsequent proteomic analysis. An amount of 75 µL of 6 M urea in 25 mM ammonium bicarbonate was added to each of the 25 µL aliquots (each aliquot contained approximately 50 µg of total protein as determined by the Bradford assay). After addition of 5 µL of 200 mM DTT to each aliquot, protein reduction was carried out for 1 h at room temperature. Protein alkylation was accomplished by adding 20 µL of 200 mM iodoacetamide and allowing it to react for 1 h at room temperature in the dark. To eliminate excess iodoacetamide in the sample prior to enzymatic digestion, 20 µL of 200 mM DTT was added. The samples were then diluted with 900 µL of 50 mM ammonium bicarbonate whereupon 2 µL of 1 µg/µL sequencing grade trypsin (Promega, Madison, WI) was added. The ER protein samples were digested overnight (18 h) at 37 °C. The enzymatic reaction was terminated the next morning upon addition of 5 µL of formic acid, and the corresponding ER protein digest was then loaded onto a Supelco C18 solid phase extraction column (Bellefonte, PA). Samples were washed several times with 0.1% TFA and eluted with 2 × 300 µL of 80% acetonitrile, 0.1% TFA. Six of the desalted ER tryptic digest samples were pooled (corresponding to 300 µg), centrifuged under vacuum until dryness, and resuspended in 100 µL of 10 mM ammonium formate, pH 3.0, containing 25% acetonitrile for subsequent SCX fractionation. One desalted digest sample was used directly for shotgun-based proteomic analysis. Strong Cation Exchange Fractionation. SCX chromatographic separation of the ER protein digest was carried out using a Polysulfethyl Aspartamide 2.1 mm i.d. x 10 cm column (PolyLC, Columbia, MD). Gradient flow rates of 200 µL/min were provided by a Surveyor (Thermo, San Jose, CA) HPLC pump where peptide elution was monitored by UV absorbance at 280 nm. Solvent A was 10 mM ammonium formate, pH 3.0, containing 25% acetonitrile, and solvent B was 500 mM ammonium formate, pH 6.8, containing 25% acetonitrile. After column equilibration in 0% solvent B, the sample was injected. Following isocratic solvent delivery for 10 min to rid the sample of nonionic impurities, a linear gradient was carried out for 30 min to 20% solvent B, ramped to 100% solvent B in 10 min, and held for an additional 10 min at 100% solvent B. Twominute fractions were automatically collected via a Foxy Jr. fraction collector (Teledyne Isco, Lincoln, NE). Eighteen SCX fractions were selected based on the UV absorbance profile and subsequently centrifuged under vacuum until dryness. Each fraction was then solubilized in 30 µL of aqueous medium containing 3% acetonitrile and 1% acetic acid for RP-HPLC tandem mass spectrometric analysis. Liquid Chromatography–Mass Spectrometry. Online RPHPLC-tandem mass spectrometric analysis of mouse brain ER tryptic digests was performed using a hybrid linear ion trap-Fourier transform ion cyclotron resonance (7-T) mass spectrometer (LTQ-FT, ThermoFisher, San Jose, CA) equipped with an electrospray ionization (ESI) source supplied by the manufacturer (NanoSpray II) and operated with the Xcalibur (version 2.2) data acquisition software. An amount of 5 µL of ER protein digest was loaded onto a PepMap C18 capillary trap (LCPackings, Sunnyvale, CA) and desalted with 3% acetonitrile and 1% acetic acid for 5 min prior to injection onto a 75 µm i.d. x 10 cm PicoFrit C18 analytical column (New Objective, Woburn, MA). Following peptide desalting and injection onto the analytical column, a linear gradient provided by a Surveyor MS pump (Thermo) was carried out to 40% acetonitrile in 90
Proteomic Analysis of Mouse Brain Microsomes
research articles
Figure 3. Comparison of overrepresented (p < 0.05) biological processes and molecular functions for the 2D-LC (black) and shotgun (gray) analysis using Ingenuity Pathway Analysis 5.0 and displayed as (a) number of proteins identified or (b) significance value for each process/function.
min at 250 nL/min. For shotgun-based analysis of the entire mouse brain ER tryptic digest (without SCX fractionation), a linear gradient was carried out to 40% acetonitrile in 60 min. Spray voltage and capillary temperature during the gradient run were maintained at 2.0 kV and 250 °C, respectively. The conventional data-dependent mode of acquisition was utilized in which an accurate m/z survey scan was performed in the FTICR cell followed by parallel MS/MS linear ion trap analysis of the top 5–10 most intense precursor ions. FTICR full-scan mass spectra were acquired at 50 000 or 100 000 mass resolving power (m/z 400) from m/z 350 to 1500 using the automatic gain control mode of ion trapping (500 000 target ion count). Collision-induced dissociation (CID) was performed in the linear ion trap using a 2.0-u isolation width and 35% normalized collision energy with helium as the target gas. Database Searching. MS/MS data generated by data-dependent acquisition via the LTQ-FT were extracted by BioWorks version 3.3 and searched against a composite IPI mouse (version 3.28, 53847 × 2 entries) protein database containing both forward and randomized sequences using the Mascot (version 2.2.1; Matrix Science, Boston, MA) search algorithm. Mascot was searched with a fragment ion mass tolerance of 0.80 Da and a parent ion tolerance of 10 ppm assuming the digestion enzyme trypsin with the possibility of one missed cleavage. Carbamidomethylation of cysteine was specified as a static modification, while oxidation of methionine, N-terminal protein acetylation, and phosphorylation of serine, threonine, and tyrosine were specified as variable modifications in the database search. The software program Scaffold (version Scaf-
fold-01_06_13, Proteome Software Inc., Portland, OR) was then employed to compile and validate tandem MS-based peptide and protein identifications. Peptide identifications were accepted at greater than 95.0% probability as determined by the Peptide Prophet algorithm.38 Protein identifications including single peptide-based identifications, where protein probabilities were assigned by the Protein Prophet algorithm,39 were accepted at greater than 80.0% probability. These filtering criteria established an acceptable false positive identification rate [(number of randomized peptide sequences identified/total number of peptides identified) × 100%] of 0.4% for the mouse brain ER proteome data set. Adjustment of the protein probability cutoff value to 95% established a peptide false positive identification rate of 0.04%; however, an approximate 30% decrease in potential true positive single peptide-based protein identifications was observed. Bioinformatics Analysis. Cytoscape equipped with the BiNGO plug-in (http://www.psb.ugent.be/cbd/papers/BiNGO/) was used to determine statistically overrepresented gene ontology (GO) categories from the mouse brain ER proteome data set. Parameters selected in the analysis included hypergeometric statistical test, biological process or cellular localization GO terms, significance level of p < 0.05, and correction for multiple term testing by Benjamini and Hochberg false discovery rate corrections. GO annotations from GOA Mouse (version 38.0, available at http://www.ebi.ac.uk/GOA/) that contained 173 476 total associations and annotation for 32 068 distinct proteins were used to create a reference set for the IPI mouse v3.28 protein database. Separate searches were perJournal of Proteome Research • Vol. 7, No. 3, 2008 1049
research articles
Stevens et al.
Figure 4. Interaction network involved in calcium signaling and transport identified from the ER proteome data set.
formed to determine overrepresented cellular localizations and biological processes associated with the ER proteins identified from the tandem MS analysis. Ingenuity Pathway Analysis 5.0 (Ingenuity Systems, Redwood City, CA) was also utilized to compare ER proteome data sets from the 2D-LC-MS/MS and shotgun-based analyses as well as to determine potential protein interaction networks. Molecular functions and biological processes were accepted at the significance level of p < 0.05 (right-tailed Fisher’s exact test).
Results and Discussion Mouse Brain ER Proteome. After subcellular fractionation to obtain mouse brain ER microsomes, a tryptic digest was performed, and the resulting peptides were fractionated by SCX chromatography. Figure 1a shows a representative UV trace monitored at 280 nm for the ER-derived tryptic digest. The gradient for the SCX analysis was optimized to generate approximately 15 two-minute fractions. Upon LC-MS/MS analysis of the SCX fractions such as fraction 25 displayed in Figure 1b, 300–400 proteins were typically identified in a single fraction. Compilation of the proteins identified from each SCX fraction resulted in the total identification of 1929 proteins from the mouse brain ER that corresponded to a relatively low false positive rate (FPR) of 0.4%. Fifteen MS/MS spectra from the entire MS data set that represented single peptide-based 1050
Journal of Proteome Research • Vol. 7, No. 3, 2008
protein identifications and appeared in various SCX fractions were discarded upon manual inspection, if annotated MS/MS spectra did not follow conventional CID fragmentation pathways. This manual filtration process essentially lowered the FPR below 0.4%, leaving a high-confidence protein catalogue containing 1914 proteins for the mouse brain ER. A complete list of the ER proteins (including potential contaminants from the organelle enrichment procedure) is provided as Supporting Information. GO Analysis of the ER Proteome. Out of 1914 proteins, 1473 could be associated with known GO terms for cellular component. Fifty-seven percent (777) of the annotated proteins were determined to be membrane proteins, thus demonstrating the effectiveness of the reported approach for the identification of hydrophobic species in contrast to 2D gel electrophoresis in which this particular protein class is typically underrepresented. Figure 2 displays specific organelle localization categories identified which included a high representation of mitochondrial and plasma membrane proteins. The presence of mitochondrial and plasma membrane proteins is not surprising since synthesis and transport of these proteins occur in the ER. Those identified could represent mitochondrial or plasma membrane-derived proteins exhibiting high turnover rates in the cell. A fair representation of ER as well as vesicle and Golgi apparatus proteins that are involved in protein transport and
Proteomic Analysis of Mouse Brain Microsomes
research articles
Figure 5. Representative MS/MS spectra of tryptic peptides derived from intracellular calcium transporters and channels in the ER including (a) sarcoplasmic/endoplasmic reticulum calcium ATPase 2, (b) isoform 1 of inositol 1,4,5-trisphosphate receptor type 1, and (c) ryanodine receptor 2.
the secretory pathway was also observed. Overrepresented biological processes determined from GO analysis of the ER proteome data set that included primarily establishment of localization (p-value of 6.5 · 10-48) and cellular component organization and biogenesis (p-value of 6.7 · 10-26) are presented in Table 1. The subcategories of these processes with known functions relevant to the ER include intracellular protein transport, the secretory pathway, as well as ER to Golgimediated vesicle transport. Other biological processes associated with cellular metabolism were identified as well, which could be due to representation of other organelle-specific proteins that were present in the ER such as mitochondrial proteins; however, their potential function in the ER cannot be precluded since various proteins can exhibit multiple organelle localizations. The presence of these metabolismrelated proteins in the ER could serve to drive certain detoxification and biosynthesis reactions. GO Comparison of 2D-LC-MS/MS vs a Shotgun-Based Approach. A shotgun-based approach was also evaluated in which an unfractionated ER tryptic digest was analyzed using a relatively short (60 min gradient) LC-MS/MS method. After filtering the data set to establish a false positive rate of
